<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <title>2!=2</title>
    <link>https://chocobear.tistory.com/</link>
    <description>생각날 때마다 포스팅하는 블로그</description>
    <language>ko</language>
    <pubDate>Thu, 21 May 2026 03:53:33 +0900</pubDate>
    <generator>TISTORY</generator>
    <ttl>100</ttl>
    <managingEditor>초코맛 도비</managingEditor>
    <image>
      <title>2!=2</title>
      <url>https://tistory1.daumcdn.net/tistory/3850060/attach/f9e9e3a45199419c99b7173e20b443e3</url>
      <link>https://chocobear.tistory.com</link>
    </image>
    <item>
      <title>추상대수학, 그 스물일곱 번째 이야기  |  가환 모노이드의 Grothendieck Group ( The Grothendieck Group of A Commutative Monoid )</title>
      <link>https://chocobear.tistory.com/211</link>
      <description>&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 가환 모노이드의 Grothendieck group을 소개하려 한다. 가환 모노이드의 Grothendieck group은 해당 가환 모노이드를 포함하는 가장 최적의 아벨군이다. 쉽게 말해서, 어떤 가환 모노이드가 주어질 때, 해당 모노이드에 역원 연산을 부여하여 얻어낸 아벨군을 해당 모노이드의 Grothendieck group이라고 한다. 실제로는 보다 엄밀한 방식으로 정의되며, 앞의 설명은 보다 직관적인 이해를 위한 설명이다. 그럼 아래의 정의를 보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, we will get to know what the Grothendieck group of a commutative monoid is. The Grothendieck group of a commutative monoid is the most general abelian group which contains the monoid. In other words, the Grothendieck group of the given commutative monoid is an abelian group obtained by introducing additive inverses. In fact, it is defined in a more rigorous way, and the preceding description is for a more intuitive understanding. Look at the below.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;b&gt;Definition 1. &lt;span class=&quot;ko&quot;&gt;가환 모노이드의 Grothendieck Group ( &lt;/span&gt;The Grothendieck Group of A Commutative Monoid&lt;span class=&quot;ko&quot;&gt; )&lt;/span&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $M$ be a commutative monoid. Its &lt;b&gt;Grothendieck group&lt;/b&gt; $K(M)$ is an abelian group with the following universal property:&lt;br /&gt;There exists a monoid homomorphism $$ \iota : M \to K(M) $$ such that for any monoid homomorphism $$ f : M \to A $$ from the commutative monoid $M$ to an abelian group $A$, there is a unique group homomorphism $$ g : K(M) \to A $$ such that $$ f = g \circ \iota. $$ This expresses the fact that any abelian group $A$ that contains a homomorphic image of $M$ will also contain a homomorphic image of $K(M)$, $K(M)$ being the &quot;most general&quot; abelian group containing a homomorphic image of $M$.&amp;nbsp;&lt;br /&gt;In addition, the map $\iota$ in the definition above is called a canonical map from $M$ to its Grothendieck group.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;가환 모노이드 $M$이 주어졌다고 하자. 이때, $M$의 &lt;b&gt;Grothendieck Group&lt;/b&gt; $K(M)$은 다음 보편 성질을 만족하는 아벨군으로 정의된다.&lt;br /&gt;가환 모노이드 $M$에서 아벨군 $A$로 가는 monoid homomorphism $ f : M \to A $가 주어질 때, 각 $f$에 대하여 $f = g \circ \iota$를 만족하는 group homomorphism $g : K(M) \to A$가 유일하게 존재하도록 하는 monoid homomorphism $\iota : M \to K(M)$가 존재한다.&lt;br /&gt;이는 $M$의 homomorphic image를 포함하는 임의의 아벨군 $A$가 언제나 $K(M)$의 homomorphic image를 포함한다는 사실을 의미하며, 이는 곧 $K(M)$이 $M$을 포함하는 최적의 아벨군임을 의미한다.&lt;br /&gt;또한, 위의 정의에서 $\iota$를 $M$에서 $K(M)$으로 가는 canonical map이라고 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we've defined the Grothendieck group of a commutative monoid, let's figure out whether the Grothendieck group always exists. Look at the below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 가환 모노이드의 Grothendieck group을 정의했으므로 Grothendieck group이 항상 존재하는지에 대해 알아보자. 아래의 정리를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;For any commutative monoids $M$, there exists a Grothendieck group $K(M)$ of $M$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;임의의 가환 모노이드 $M$에 대하여, $M$의 Grothendieck group $K(M)$이 항상 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $F_{\operatorname{ab}}(M)$ be the free abelian group generated by $M$. We denote the generator of $F_{\operatorname{ab}}(M)$ corresponding to an element $x \in M$ by $\left[ x \right]$. Let $B$ be the subgroup generated by all elements of type $$ \left[ x+y \right] - \left[ x \right] - \left[ y \right] $$ where $x,y \in M$. We let $K(M) = F_{\operatorname{ab}}(M)/B$, and let $$ \iota : M \to K(M) $$ be the map obtained by composing the injection $\gamma$ of $M$ into $F_{\operatorname{ab}}(M)$ given by $\gamma : x \mapsto \left[ x \right]$, and the canonical map $$ \pi : F_{\operatorname{ab}}(M) \to F_{\operatorname{ab}}(M)/B. $$ It is then clear that $$ \iota(x)+\iota(y) = \pi(\left[ x \right])+\pi(\left[ y \right]) = \pi(\left[ x \right] + \left[ y \right]) = \pi(\left[ x+y \right]) = \iota(x+y) $$ since $\left[ x+y \right] - \left[ x \right] - \left[ y \right] \in B$. Hence $\iota$ is a homomorphism.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;By &lt;a href=&quot;https://chocobear.tistory.com/210#Theorem3&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;the universal property of free abelian group&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;, $\iota$ obviously satisfies the desired universal property and the theroem is proved.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$M$으로부터 생성된 자유 아벨군을 $F_{\operatorname{ab}}(M)$으로 표기하고 각 $x \in M$에 대응되는 $F_{\operatorname{ab}}(M)$의 자유 생성자를 $\left[ x \right]$로 표기하도록 하자. 이제 $B$를 $\left[ x+y \right] - \left[ x \right] - \left[ y \right]$ 꼴의 원소들로부터 생성된 $F_{\operatorname{ab}}$의 부분군이라고 하자. 이제 $K(M) = F_{\operatorname{ab}}(M)/B$라고 하고 $\iota : M \to K(M)$를 $\gamma : \begin{cases} M \to F_{\operatorname{ab}}(M) \\ x \mapsto \left[ x \right] \end{cases}$와 canonical map $\pi : F_{\operatorname{ab}}(M) \to F_{\operatorname{ab}}(M)/B$의 합성으로 정의되는 사상이라고 하자. 그러면 $\left[ x+y \right] - \left[ x \right] - \left[ y \right] \in B$라는 사실로부터 $$ \iota(x) + \iota(y) = \pi(\left[ x \right]) + \pi(\left[ y \right]) = \pi(\left[ x \right] + \left[ y \right]) = \pi(\left[ x+y \right]) = \iota(x+y) $$임을 알 수 있고, 따라서 $\iota$가 homomorphism임을 알 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/210#Theorem3&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;자유 아벨군의 보편 성질&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 의해, $\iota$가 목표했던 보편 성질을 만족한다는 사실을 알 수 있으며, 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위의 정리를 통해 우리는 Grothendieck group을 구성하는 한 가지 방법을 알 수 있다. 위와 같은 방법 외에도 Grothendieck group을 구성하는 방법이 한 가지 더 있다. 아래를 보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;By the above, we can know one way to construct a Grothendieck group of the given commutative monoid. In addition to the above method, there is one more way to construct a Grothendieck group of the given commutative monoid. Look at the below.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem2&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $M$ be a commutative monoid and define the addition on $M \times M$ component-wise: $$ \left( m_1, m_2 \right) + \left( m'_1, m'_2 \right) = \left( m_1 + m'_1, m_2 + m'_2 \right) $$ Now we shall define a binary relation $\sim$ on $M \times M$ as $\left( m_1, m_2 \right) \sim \left( m'_1, m'_2 \right)$ if and only if there exists a $c \in M$ such that $m_1 + m'_2 + c = m_2 + m'_1 + c$. Then the relation $\sim$ is an &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/19&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;equivalence relation&lt;/b&gt;&lt;/a&gt; and $\left( M \times M \right) / \sim$ is a Grothendieck group of $M$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;가환 모노이드 $M$이 주어졌다고 하고, $M \times M$ 위의 덧셈을 성분별 덧셈으로 정의하자. 즉, $M \times M$ 위의 덧셈을 아래와 같이 정의하자. $$ \left( m_1, m_2 \right) + \left( m'_1, m'_2 \right) = \left( m_1 + m'_1, m_2 + m'_2 \right) $$ 이제 $m_1 + m'_2 + c = m_2 + m'_1 + c$이도록 하는 $c \in M$가 존재하는 경우를 $\left( m_1, m_2 \right) \sim \left( m'_1, m'_2 \right)$로 정의하자. 그러면 $M \times M$ 위의 이항 관계 $\sim$은 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/19&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;동치관계&lt;/b&gt;&lt;/a&gt;이며, $\left( M \times M \right) / \sim$은 $M$의 Grothendieck group이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;First, we shall prove that $\sim$ is an equivalence relation on $M \times M$. It is clear that $\sim$ is reflective and symmetric. Now let $(a_1, b_1) \sim (a_2, b_2)$ and $(a_2, b_2) \sim (a_3, b_3)$. Then there exist two elements $c_{12}, c_{23} \in M$ such that $a_1+b_2+c_{12} = a_2+b_1+c_{12}$ and $a_2+b_3+c_{23} = a_3+b_2+c_{23}$. Then it is clear that $a_1+b_3+a_2+b_2+c_{12}+c_{23} = a_3+b_1+a_2+b_2+c_{12}+c_{23}$ and thus makes $(a_1, b_1) \sim (a_3, b_3)$. Therefore $\sim$ is transitive and, hence, $\sim$ is an equivalence relation on $M \times M$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now it is left to show that $\left( M \times M \right) / \sim$ is a Grothendieck group of $M$. Since $M \times M$ is a commutative monoid, it is clear that $\left( M \times M \right) / \sim$ satisfies the closure property under the addition and the associativity and commutativity of the addition. Also it is trivial that $\left[ \left( 0,0 \right) \right]_{\sim}$ is an identity element of $\left( M \times M \right) / \sim$. Since $\left( m,m \right) \sim \left( 0,0 \right)$ for any elements $m \in M$, $\left[ \left( m_1, m_2 \right) \right]_{\sim} + \left[ \left( m_2, m_1 \right) \right]_{\sim} = \left[ \left( 0,0 \right) \right]_{\sim}$. So every element of $\left( M \times M \right) / \sim$ has an inverse element and therefore $\left( M \times M \right) / \sim$ is an abelian group.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We shall define a monoid homomorphism $\iota : M \to \left( M \times M \right) / \sim$ as $m \mapsto \left[ \left( 0,m \right) \right]_{\sim}$. Now let an abelian group $A$ and a monoid homomorphism $f : M \to A$ be the given. We shall show that $\left[ \left( 0,m \right) \right]_{\sim} = \left[ \left( 0,m' \right) \right]_{\sim}$ implies that $f(m) = f(m')$. By the definition of the relation $\sim$, there exists an element $c \in M$ such that $m+c = m'+c$. So it is trivial that $f(m) = f(m) + f(c) - f(c) = f(m+c) - f(c) = f(m'+c) - f(c) = f(m') + f(c) - f(c) = f(m')$. Hence, there exists a group homomorphism $g : \left( M \times M \right) / \sim \to A$ such that $f = g \circ \iota$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let a group homomorphism $g' : \left( M \times M \right) / \sim \to A$ also satisfy the condition that $f = g' \circ \iota$. Then for any $m \in M$, $g \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} \right)$ and thus makes $g \left( \left[ \left( m,m' \right) \right]_{\sim} \right) = g \left( \left[ \left( 0,m' \right) \right]_{\sim} - \left[ \left( 0,m \right) \right]_{\sim} \right) = g \left( \left[ \left( 0,m' \right) \right]_{\sim} \right) - g \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} \right) - g' \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} - \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( m,m' \right) \right]_{\sim} \right)$. Hence, $g = g'$ and $\left( M \times M \right) / \sim$ is therefore a Grothendieck group of $M$ and $\iota$ is a canonical map of $M$ into $\left( M \times M \right) / \sim$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;먼저, $\sim$이 $M \times M$ 위의 동치관계임을 보이자. 먼저, $\sim$의 정의로부터 $\sim$이 반사적이고 대칭적이라는 것은 매우 자명하다. 이제 $( a_1, b_1 ) \sim ( a_2, b_2 )$이고 $( a_2, b_2 ) \sim ( a_3, b_3 )$이라고 하자. 그러면 $\sim$의 정의에 의해 $a_1 + b_2 + c_{12} = a_2 + b_1 + c_{12}$와 $a_2 + b_3 + c_{23} = a_3 + b_2 + c_{23}$이 성립하도록 하는 $c_{12}, c_{23} \in M$이 존재한다. 이때, 두 등식의 양변을 서로 더해주면, $a_1 + b_3 + ( a_2 + b_2 + c_{12} + c_{23} ) = a_3 + b_1 + ( a_2 + b_2 + c_{12} + c_{23} )$이 성립하므로 $( a_1 , b_1 ) \sim ( a_3 , b_3 )$임을 알 수 있다. 따라서 $\sim$은 추이관계임을 알 수 있으며, 이로 인해 $\sim$이 동치관계임이 증명된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $\left( M \times M \right) / \sim$이 $M$의 Grothendieck group이라는 것을 보이는 것만이 남았다. $M \times M$이 가환 모노이드라는 사실로부터, $\left( M \times M \right) / \sim$이 덧셈에 대해 닫혀 있으며, 덧셈에 대한 교환법칙과 결합법칙을 만족한다는 것을 알 수 있다. 또한, $\left[ \left( 0 , 0 \right) \right]_{\sim}$이 항등원이 된다는 사실이 매우 자명하며, $\left( M \times M \right) / \sim$의 임의의 원소 $\left[ \left( a , b \right) \right]_{\sim}$에 대하여 $\left[ \left( a , b \right) \right]_{\sim} + \left[ \left( b , a \right) \right]_{\sim} = \left[ \left( a , b \right) \right]_{\sim}$임을 쉽게 알 수 있으므로, $\left( M \times M \right) / \sim$이 아벨군임이 보여진다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 monoid homomorphism $\iota : M \to \left( M \times M \right) / \sim$를 $\iota : m \mapsto \left[ \left( 0 , m \right) \right]_{\sim}$으로 정의하고 아벨군 $A$와 monoid homomorphism $f : M \to A$가 주어졌다고 하자. 이제 $\left[ \left( 0 , m \right) \right]_{\sim} = \left[ \left( 0 , m' \right) \right]_{\sim}$이 $f(m) = f(m')$을 함의한다는 것을 보일 것이다. $\sim$의 정의에 의해 $m + c = m' + c$가 성립하도록 하는 $c \in M$가 존재한다. 따라서 다음이 성립한다. $$ f(m) = f(m) + f(c) - f(c) = f(m+c) - f(c) = f(m'+c) - f(c) = f(m') + f(c) - f(c) = f(m') $$ 따라서 $f = g \circ \iota$를 만족하도록 하는 group homomorphism $g : \left( M \times M \right) / \sim \to A$의 존재성은 자명하다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $f = g \circ \iota = g' \circ \iota$인 두 group homomorphism $g : \left( M \times M \right) / \sim \to A$와 $g' : \left( M \times M \right) / \sim \to A$가 존재한다고 하자. 그러면 임의의 $m \in M$에 대하여 $g \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} \right)$가 성립하며, 따라서 $g \left( \left[ \left( m,m' \right) \right]_{\sim} \right) = g \left( \left[ \left( 0,m' \right) \right]_{\sim} - \left[ \left( 0,m \right) \right]_{\sim} \right) = g \left( \left[ \left( 0,m' \right) \right]_{\sim} \right) - g \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} \right) - g' \left( \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( 0,m' \right) \right]_{\sim} - \left[ \left( 0,m \right) \right]_{\sim} \right) = g' \left( \left[ \left( m,m' \right) \right]_{\sim} \right)$가 성립한다. 따라서 두 group homomorphism $g$와 $g'$은 같은 함수이며, 이로 인해 $\left( M \times M \right) / \sim$이 $M$의 Grothendieck group이며 $\iota$가 $\left( M \times M \right) / \sim$으로 가는 $M$의 canonical map이 됨을 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we know two different ways to construct a Grothendieck group of the given commutative monoid, you might want to know whether these two Grothendieck groups of the given commutative monoid obtained by the two methods above are isomorphic to each other. To answer to the question, we need a lemma. Look at the below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;가환 모노이드의 Grothendieck group을 구성하는 두 가지 방법을 알게 되었으므로, 이 두 가지 방법을 통해 얻은 두 Grothendieck group이 isomorphic한지가 궁금할 것이다. 이 질문에 대한 답을 하기 위해서 우리는 한 가지 보조정리가 필요하다. 아래를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Lemma1&quot;&gt;&lt;/a&gt;&lt;b&gt;Lemma 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr class=&quot;en&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;Let $M$ be a commutative monoid, $K(M)$ be a Grothendieck group of $M$, and $\iota$ be a canonical map of $M$ into $K(M)$. Then $K(M) = \{ a-b \;|\; a,b \in \iota(M) \}$.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr class=&quot;ko&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;$M$이 가환 모노이드라 하고, $K(M)$을 $M$의 Grothendieck group, $\iota$를 $K(M)$으로 가는 $M$의 canonical map이라고 하자. 그러면 $K(M) = \{ a-b \;|\; a,b \in \iota(M) \}$이다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $B = \{ a-b \;|\; a,b \in \iota(M) \}$. Then it is obvious that $B$ is a subgroup of $K(M)$. Since $K(M)$ is an abelian group, $B$ is a normal subgroup of $K(M)$. If $B \neq K(M)$, then $K(M)/B$ is not a trivial group and then there exist more than one group homomorphisms $f : K(M)/B \to K(M)/B$. Since $K(M) \cong (K(M)/B) \times B$, there exist more than one group homomorphism $h : K(M) \to A$ such that $g = h \circ \iota$ where a monoid homomorphism $g : M \to A$ is given and $A$ is an abelian group. This is obviously a contradiction the fact that $K(M)$ is a Grothendieck group of $M$ and $\iota$ is a canonical map of $M$ into $K(M)$. Therefore, $B = K(M)$ and, hence, the theorem is proved.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$B = \{ a-b \;|\; a,b \in \iota(M) \}$라고 하자. 그러면 $B$가 $K(M)$의 부분군임은 자명하다. $K(M)$이 아벨군이므로, $B$는 $K(M)$의 정규부분군임을 알 수 있다. 만약 $B \neq K(M)$이라면, $K(M)/B$가 자명군이 아니게 되므로 $K(M)/B \to K(M)/B$인 group homomorphism이 적어도 두 개 있다. 이때, $K(M) \cong (K(M)/B) \times B$라는 사실로부터 아벨군 $A$와 monoid homomorphism $g : M \to A$가 주어질 때, $g = h \circ \iota$를 만족하도록 하는 group homomorphism $h : K(M) \to A$가 적어도 두 개 있음을 알 수 있으며, 이는 $K(M)$이 $M$의 Grothendieck group이고 $\iota$가 $K(M)$으로 가는 $M$의 canonical map이라는 사실에 모순이다. 따라서 $B = K(M)$이며, 이로 인해 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 위의 질문에 대해 답을 할 차례이다. 아래를 보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now, it is time to answer the question above. Look at below.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem3&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr class=&quot;en&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;Let $M$ be a commutative monoid and both $K(M)$ and $K'(M)$ be Grothendieck Groups of $M$. Then $K(M)$ is isomorphic to $K'(M)$.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr class=&quot;ko&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;가환 모노이드 $M$과 그의 Grothendieck group $K(M)$과 $K'(M)$이 주어졌다고 하자. 그러면 $K(M)$과 $K'(M)$은 isomorphic하다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $K(M)$ and $K'(M)$ be two Grothendieck groups of $M$ and $\gamma$ and $\gamma'$ be canonical map of $M$ into $K(M)$ and $K'(M)$, respectively. Then by definition, there exists a unique homomorphism $\gamma_* : K'(M) \to K(M)$ making the following diagram commutative:&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$K(M)$과 $K'(M)$이 $M$의 두 Grothendieck group이라고 하고, $\gamma$와 $\gamma'$을 각각 $K(M)$과 $K'(M)$으로 가는 $M$의 canonical map이라고 하자. 그러면 정의에 의해 다음 다이어그램이 가환이도록 하는 homomorphism $\gamma_* : K'(M) \to K'(M)$이 유일하게 존재한다.&lt;/p&gt;
&lt;p&gt;&lt;iframe src=&quot;https://q.uiver.app/?q=WzAsMyxbMCwwLCJNIl0sWzIsMCwiSycoTSkiXSxbMSwyLCJLKE0pIl0sWzAsMiwiXFxnYW1tYSIsMl0sWzAsMSwiXFxnYW1tYSciXSxbMSwyLCJcXGdhbW1hXyoiXV0=&amp;amp;embed&quot; width=&quot;486&quot; height=&quot;432&quot;&gt;&lt;/iframe&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Define $\gamma'_*$ similarly. Then the following diagram is commutative:&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\gamma'_*$ 역시 $\gamma_*$와 비슷하게 정의하자. 그러면 아래의 다이어그램이 가환이다.&lt;/p&gt;
&lt;p&gt;&lt;iframe src=&quot;https://q.uiver.app/?q=WzAsMyxbMCwwLCJNIl0sWzIsMCwiSycoTSkiXSxbMSwyLCJLKE0pIl0sWzAsMiwiXFxnYW1tYSIsMl0sWzAsMSwiXFxnYW1tYSciXSxbMiwxLCJcXGdhbW1hJ18qIiwyXV0=&amp;amp;embed&quot; width=&quot;486&quot; height=&quot;432&quot;&gt;&lt;/iframe&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Hence, $(\gamma_* \circ \gamma'_*) (x) = x$ for every element $x \in \gamma(M)$ and $(\gamma'_* \circ \gamma_*) (x) = x$ for every element $x \in \gamma'(M)$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $\gamma_*$ and $\gamma'_*$ are group homomorphisms, $\{ a-b \;|\; a,b \in \gamma(M) \} \cong \{ a-b \;|\; a,b \in \gamma'(M) \}$. Therefore, by the &lt;a href=&quot;#Lemma1&quot;&gt;&lt;b&gt;Lemma 1&lt;/b&gt;&lt;/a&gt;, $K(M) \cong K'(M)$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 임의의 $x \in \gamma(M)$에 대하여 $(\gamma_* \circ \gamma'_*) (x) = x$이고, 임의의 $x \in \gamma'(M)$에 대하여 $(\gamma'_* \circ \gamma_*) (x) = x$임을 알 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\gamma_*$와 $\gamma'_*$가 group homomorphism이므로 $\{ a-b \;|\; a,b \in \gamma(M) \} \cong \{ a-b \;|\; a,b \in \gamma'(M) \}$이다. 따라서, &lt;a href=&quot;#Lemma1&quot;&gt;&lt;b&gt;Lemma 1&lt;/b&gt;&lt;/a&gt;에 의해, $K(M) \cong K'(M)$이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;또한, 위 정리의 증명과정을 보면, 가환 모노이드 $M$의 두 Grothendieck group $K(M)$, $K'(M)$ 각각으로 가는 $M$의 canonical map $\gamma$와 $\gamma'$에 대하여 $\gamma = f \circ \gamma'$가 성립하도록 하는 group homomorphism $f : K'(M) \to K(M)$이 유일하게 존재한다는 사실을 추가적으로 알 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이 외에도 위의 정리에 의한 결과로 다음과 같은 따름정리를 얻을 수 있다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Furthermore, by the procedure of the proof above, we can easily verify that two canonical maps $\gamma$ and $\gamma'$ of a commutative monoid $M$, which are into the Grothendieck groups $K(M)$ and $K'(M)$ of $M$, respectively, there uniquely exists a group homomorphism $f : K'(M) \to K(M)$ such that $\gamma = f \circ \gamma'$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Additionally, we can get the following corollary as the result of the theorem above.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Corollary3_1&quot;&gt;&lt;/a&gt;&lt;b&gt;Corollary 3.1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr class=&quot;en&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;Let $M$ be a commutative monoid, $K(M)$ be a Grothendieck group of $M$, and $\gamma : M \to K(M)$ be a canonical map of $M$. Then the followings hold:&lt;br /&gt;(a) $\gamma$ is injective if and only if $M$ has the cancellation property,&lt;br /&gt;(b) and $\gamma$ is bijective if and only if $M$ is already a group.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr class=&quot;ko&quot;&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;가환 모노이드 $M$과 그의 Grothendieck group $K(M)$, $M$의 canonical map $\gamma : M \to K(M)$이 주어졌다고 하자. 그러면 다음이 성립한다.&lt;br /&gt;(a) $\gamma$가 단사인 것과 $M$이 소거법칙을 만족하는 것은 동치이다.&lt;br /&gt;(b) $\gamma$가 전단사인 것과 $M$이 그 자체로 군이 되는 것은 동치이다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위 따름정리는 비교적 자명하므로 독자들에게 맡긴다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since the corollary above is quite obvious, the proof is left to the readers.&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Grothendieck Group</category>
      <category>Grothendieck Group of Commutative Monoid</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/211</guid>
      <comments>https://chocobear.tistory.com/211#entry211comment</comments>
      <pubDate>Sun, 20 Mar 2022 03:54:15 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물여섯 번째 이야기  |  자유 아벨군 ( Free Abelian Group )</title>
      <link>https://chocobear.tistory.com/210</link>
      <description>&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, we will learn about what free abelian group is. Free abelian group is an abelian group which has a basis. If you learned linear algebra, you're probably familiar with the concept of basis. The concept of basis in group theory is similar to the concept of basis in linear algebra. So if you studied linear algebra, you'll easily understand the concept of free abelian group. First, look at the definition below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 자유 아벨군에 대한 이야기를 해보려 한다. 자유 아벨군은 기저를 갖는 아벨군을 말한다. 만약 지금 이 글을 읽고 있는 여러분이 선형대수를 공부해본 적이 있다면 아마 꽤 높은 확률로 기저의 개념에 익숙할 것이다. 군론에서의 기저의 개념 역시 선형대수에서의 그것과 크게 다르지 않다. 따라서 선형대수를 공부해본 사람이라면 이번 글을 좀 더 쉽게 이해할 수 있을 것이다. 그럼 이제 아래의 정의를 살펴보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. Direct Sum&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $\{ A_i \}_{i \in I}$ be a family of abelian groups. We define their &lt;b&gt;direct sum&lt;/b&gt; $$ A = \underset{i \in I}{\bigoplus} A_i $$ to be the subset of the direct product $\prod A_i$ consisting of all families $\left( x_i \right)_{i \in I}$ with $x_i \in A_i$ such that $x_i = 0$ for all but a finite number of indices $i$. Then it is clear that $A$ is a subgroup of the product. For each index $j \in I$, we map $$ \lambda_j : A_j \to A $$ by letting $\lambda_j(x)$ be the element whose $j$-th component is $x$, and having all other components equal to $0$. Then $\lambda_j$ is an injective homomorphism.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$\{ A_i \}_{i \in I}$가 아벨군의 집합족이라고 하자. 그러면 이들의 &lt;b&gt;Direct Sum&lt;/b&gt; $$ A = \underset{i \in I}{\bigoplus} A_i $$은 유한 개의 첨수를 제외한 모든 $i \in I$에 대하여 $x_i = 0$을 만족하도록 하는 $x_i \in A_i$들의 순서쌍 $\left( x_i \right)_{i \in I}$를 원소로 가지는 direct product $\prod A_i$의 부분집합으로 정의된다. 그러면 $A$가 direct product의 부분군이 됨은 매우 자명하다. 각 첨수 $j \in I$에 대하여 사상 $$ \lambda_j : A_j \to A $$를 $\lambda_j(x)$가 $j$번째 성분이 $x$이고 나머지 성분은 $0$인 원소에 대응되도록 정의하자. 그러면 $\lambda_j$는 단사 homomorphism이 됨을 알 수 있다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 1. 보편 성질 ( Universal Property )&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 1. Universal Property&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $\{ f_i : A_i \to B \}$ be a family of homomorphisms into an abelian group $B$. Let $A = \bigoplus A_i$. There exists a unique homomorphism $$ f : A \to B $$ such that $f \circ \lambda_j = f_j$ for all $j$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;아벨군 $B$로의 homomorphism의 집합 $\{ f_i : A_i \to B \}$가 주어졌다고 하고 $A = \bigoplus A_i$라고 하자. 그러면 모든 $j$에 대하여 $f \circ \lambda_j = f_j$가 되도록 하는 homomorphism $ f : A \to B $가 유일하게 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We can define a map $f : A \to B$ by the rule&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ f((x_i)_{i \in I}) = \sum_{i \in I} f_i(x_i). $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The sum on the right is actually finite since all but a finite number of terms are $0$. It is immediately verified that our map $f$ is a homomorphism. Furthermore, we clearly have $f \circ \lambda_j (x) = f_j (x)$ for each $j$ and each $x \in A_j$. Thus $f$ has the desired commutativity property. It is also clear that the map $f$ is uniquely determined, as was to be shown.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;함수 $f : A \to B$를 다음과 같이 정의하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$$ f((x_i)_{i \in I}) = \sum_{i \in I} f_i(x_i) $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;우변의 합은 유한 개의 항을 제외하고는 모두 $0$이기 때문에 사실은 유한합임을 알 수 있다. 따라서 $f$가 homomorphism임은 그 즉시 알 수 있으며, 더 나아가서 $f \circ \lambda_j (x) = f_j (x)$가 임의의 $j$와 $x \in A_j$에 대하여 성립함이 자명하다. 따라서 $f$는 목표했던 가환성을 지니며, 이러한 $f$가 유일하게 결정된다는 것 역시 매우 자명하므로 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The property expressed in&amp;nbsp;&lt;b&gt;Theorem 1&lt;/b&gt; is called the&amp;nbsp;&lt;b&gt;universal property&lt;/b&gt; of the direct sum. If you want to know more about universal property, google what universal property is. If the index set $I$ is a finite set in the definition of the direct sum, depending on the context, the following definitions may be used.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위 정리에서 설명된 성질은 direct sum의 보편 성질이라고 불린다. 보편 성질에 대해 더 깊게 알고 싶다면 직접 찾아보는 걸 추천한다. 만약 direct sum의 정의에서 첨수 집합인 $I$가 유한 집합이라면, 문맥에 따라 아래의 정의를 사용하기도 한다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition2&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $A$ be an abelian group and $B$, $C$ subgroups. If $B+C = A$ and $B \cap C = \{ 0 \}$ then the map $$B \times C \to A$$ given by $\left( x , y \right) \mapsto x+y$ is an isomorphism. Instead of writing $A = B \times C$ we shall write $$ A = B \oplus C $$ and say that $A$ is the&amp;nbsp;&lt;b&gt;direct sum&lt;/b&gt; of $B$ and $C$. We use a similar notation for the direct sum of a finite number of subgroups $B_1,\cdots,B_n$ such that $$B_1 + B_2 + \cdots + B_n = A$$ and $$ B_{i+1} \cap \left( B_1 + B_2 + \cdots + B_i \right) = \{ 0 \}. $$ In that case we write $$ A= B_1 \oplus B_2 \oplus \cdots \oplus B_n. $$&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$A$가 아벨군이고 $B$와 $C$가 그의 부분군이라고 하자. 만약 $B+C = A$이고 $B \cap C = \{ 0 \}$이라면 $\left( x , y \right) \mapsto x+y$로 주어지는 사상 $$ B \times C \to A $$이 $B \times C$와 $A$ 사이의 isomorphism이 된다. 이 경우에는 $A = B \times C$라고 쓰는 대신에 $$ A = B \oplus C $$와 같이 쓰며, $A$를 $B$와 $C$의&amp;nbsp;&lt;b&gt;Direct Sum&lt;/b&gt;이라고 부른다. 이러한 노테이션은 유한 개의 부분집합 $B_1, \cdots, B_n$에 대하여 $$ B_1 + B_2 + \cdots + B_n = A $$와 $$ B_{i+1} \cap \left( B_1 + B_2 + \cdots + B_i \right) = \{ 0 \} $$이 성립하는 경우에도 $$ A = B_1 \oplus B_2 \oplus \cdots \oplus B_n $$와 같이 사용된다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition3&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 3. 기저와 자유 아벨군 ( Basis &amp;amp; Free Abelian Group )&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 3. Basis &amp;amp; Free Abelian Group&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $A$ be an abelian group. Let $\{ e_i \}_{i \in I}$ be a family of elements of $A$. We say that this family is a&amp;nbsp;&lt;b&gt;basis&lt;/b&gt; for $A$ if the family is not empty, and if every element of $A$ has a unique expression as a linear combination $$ x = \sum x_i e_i $$ with $x_i \in \mathbb{Z}$ and $x_i = 0$ for all but a finite number of indices $i$. Thus the sum is actually a finite sum. An abelian group is said to be&amp;nbsp;&lt;b&gt;free&lt;/b&gt; if it has a basis.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$A$가 아벨군이라고 하고 $\{ e_i \}_{i \in I}$가 $A$의 원소들만으로 이루어진 집합이라고 하자. 만약 이 집합이 공집합이 아니고 동시에 임의의 $A$의 원소에 대하여 해당 원소를 표현하는 $$ x = \sum x_i e_i $$꼴의 선형결합 표현이 유일할 때 이 집합 $\{ e_i \}_{i \in I}$을 $A$의 &lt;b&gt;기저 ( Basis )&lt;/b&gt;라고 부른다. 이때, 선형결합 표현에서 계수 $x_i$는 모두 $\mathbb{Z}$의 원소이며, 유한한 개수의 첨수 $i$를 제외하고는 $x_i = 0$이어야 한다. 즉, $A$의 모든 원소는 $\{ e_i \}_{i \in I}$의 원소들의 유한합으로 표현 가능해야 한다. 기저를 가지는 아벨군을 &lt;b&gt;자유 아벨군 ( Free Abelian Group )&lt;/b&gt;이라고 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem2&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $A$ be a free abelian group and $\mathcal{B} = \{ b_i \}_{i \in I}$ be a basis of $A$. Then $A \cong \displaystyle \bigoplus_{i \in I} \mathbb{Z}$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$A$가 자유 아벨군이고 $\mathcal{B} =\{ b_i \}_{i \in I}$가 $A$의 기저라고 하자. 그러면 $A \cong \displaystyle \bigoplus_{i \in I} \mathbb{Z}$이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $x \in A$ and $x = \displaystyle \sum_{i \in I} x_i b_i$ the linear combination presentation of $x$. Then $x_i = 0$ for all but a finite number of indices $i$. Thus $\left( x_i \right)_{i \in I}$ is an element of $\displaystyle \bigoplus_{i \in I} \mathbb{Z}$. Now we define a map $f : A \to \displaystyle \bigoplus_{i \in I} \mathbb{Z}$ as&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ x = \sum_{i \in I} x_i b_i \mapsto \left( x_i \right)_{i \in I}. $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Then it is immediately verified that $f$ is a homomorphism. Also it is obvious that $f$ is a one-to-one correspondence and $f$ is therefore an isomorphism. This proves the theorem.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$x \in A$라고 하고 $x = \displaystyle \sum_{i \in I} x_i b_i$가 $x$의 선형결합 표현이라고 하자. 그러면 유한 개의 첨수 $i$를 제외한 모든 $i$에 대하여 $x_i = 0$이므로 $\left( x_i \right)_{i \in I}$는 $\displaystyle \bigoplus_{i \in I} \mathbb{Z}$의 원소가 됨을 알 수 있다. 이제 사상 $f : A \to \displaystyle \bigoplus_{i \in I} \mathbb{Z}$를&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$$ x = \sum_{i \in I} x_i b_i \mapsto \left( x_i \right)_{i \in I} $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;와 같이 정의하자. 그러면 $f$가 homomorphism이 됨은 그 즉시 보여지며 $f$가 일대일 대응이라는 것 역시 자명하다. 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Next we shall define the free abelian group generated by a set. Look at below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 집합으로부터 생성된 자유 아벨군을 정의할 것이다. 아래를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 4. 집합으로부터 생성된 자유 아벨군 ( Free Abelian Group Generated by Set )&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 4. The Free Abelian Group Generated by A Set&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $S$ be a set. We shall define the free abelian group generated by $S$ as follows. Let $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ be the set of all maps $\phi : S \to \mathbb{Z}$ such that $\phi(x) = 0$ for all but a finite number of elements $x \in S$. Then $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ is an abelian group (addition being the usual addition of maps). If $k$ is an integer and $x$ is an element of $S$, we denote by $k \cdot x$ the map $\phi$ such that $\phi(x) = k$ and $\phi(y) = 0$ if $y \neq x$. Then it is obvious that every element $\phi$ of $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ can be written in the form $$ \phi = k_1 \cdot x_1 + \cdots + k_n \cdot x_n $$ for some integers $k_i$ and elements $x_i \in S$, all the $x_i$ being distinct. Furthermore, $\phi$ admits a unique expression, because if we have $$ \phi = \sum_{x \in S} k_x \cdot x = \sum_{x \in S} k'_x \cdot x $$ then $$ 0 = \sum_{x \in S} \left( k_x - k'_x \right) \cdot x, $$ whence $k'_x = k_x$ for all $x \in S$. We shall denote $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ also by $F_\mathrm{ab} (S)$, and call $F_\mathrm{ab} (S)$ the &lt;b&gt;free abelian group generated by $S$&lt;/b&gt;. We call elements of $S$ its &lt;b&gt;free generators&lt;/b&gt;.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;집합 $S$가 주어졌다고 하자. 그러면 $S$로부터 생성된 자유 아벨군을 다음과 같이 정의한다. $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 유한 개의 $x \in S$를 제외한 모든 $x$에 대하여 $\phi(x) = 0$이 성립하도록 하는 함수 $\phi : S \to \mathbb{Z}$의 집합이라고 하자. 그러면 함수의 덧셈 연산 아래에서 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$가 아벨군을 이룬다는 것을 알 수 있다. 만약 $k$가 정수이고 $x \in S$라면, $\phi(x) = k$이고 $x \neq y$이면 $\phi(y) = 0$인 함수 $\phi : S \to \mathbb{Z}$를 $k \cdot x$와 같이 나타낸다. 그러면 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$의 모든 원소 $\phi$는 몇 개의 정수들 $k_i$와 서로 다른 $x_i \in S$들에 대하여 $$ \phi = k_1 \cdot x_1 + \cdots + k_n \cdot x_n $$과 같이 표현할 수 있음이 자명하다. 또한, 이러한 표현법은 유일한데, 만약 $$ \phi = \sum_{x \in S} k_x \cdot x = \sum_{x \in S} k'_x \cdot x $$라면, $$ 0 = \sum_{x \in S} \left( k_x - k'_x \right) \cdot x $$가 성립하므로 $k_x = k'_x$가 성립한다는 것을 그 즉시 알 수 있기 때문이다. 때때로 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 $F_\mathrm{ab} (S)$와 같이 나타내기도 하며, $F_\mathrm{ab} (S)$를 $S$로부터 생성된 자유 아벨군 ( Free Abelian Group Generated by $S$ )이라고 한다. 또한, $S$의 원소들을 $F_\mathrm{ab} (S)$의 &lt;b&gt;자유 생성자 ( Free Generator )&lt;/b&gt;라고 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;It is customary to identify $S$ in $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$, and we sometimes omit the dot when we write $k_x x$ or a sum $\sum k_x x$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$가 주어질 때 $S$를 식별하는 것은 상당히 어려운 작업이며, 때때로 원소들을 나타낼 때 $k_x x$나 $\sum k_x x$와 같이 점을 생략해서 나타내기도 한다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem3&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $f_S$ be a mapping of $S$ into $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ such that $f_S(x) = 1 \cdot x$. It is then clear that $f_S$ is injective, and that $f_S(S)$ generates $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$. If $g : S \to B$ is a mapping of $S$ into some abelian group $B$, then there exists a homomorphism $g_* : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to B$ such that $g_* \circ f_S = g$ and such homomorphism is unique.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;함수 $f_S : S \to \mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 $x \mapsto 1 \cdot x$로 정의하자. 그러면 $f_S$가 단사사상이며 $f_S(S)$가 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 생성한다는 것이 자명하다. 이제 아벨군 $B$에 대하여 함수 $g : S \to B$가 주어졌다고 하자. 그러면 $g_* \circ f_S = g$를 만족하는 homomorphism $g_* : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to B$가 유일하게 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let's define $g_* : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to B$ as&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ g_* : \sum_{x \in S} k_x \cdot x \mapsto \sum_{x \in S} k_x g(x). $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;It is then clear that $g_*$ is a homomorphism and that $g_* \circ f_S = g$. Thus, such homomorphism exists and now only showing uniqueness of such homomorphism is remaining.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let's focus on the condition that $g_* \circ f_S = g$. We can rewrite $g_* \circ f_S = g$ to $g_*(1 \cdot x) = g(x)$. Since $g_*$ is a homomorphism and $f_S(S)$ generates $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$, such $g_*$ is unique, obviously.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;사상 $g_* : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to B$를 다음과 같이 정의하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$$ g_* : \sum_{x \in S} k_x \cdot x \mapsto \sum_{x \in S} k_x g(x) $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러면 $g_*$가 homomorphism이고 $g_* \circ f_S = g$가 성립함은 분명하다. 따라서 조건을 만족하는 homomorphism은 존재하며, 이제 이러한 homomorphism의 유일성만을 보이면 정리가 증명된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $g_* \circ f_S = g$라는 조건에 집중하자. $g_* \circ f_S = g$는 $g_* ( 1 \cdot x ) = g(x)$로 고쳐 쓸 수 있다. 이때 $g_*$가 homomorphism이고 $f_S(S)$가 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 생성한다는 사실로부터 조건을 만족하는 $g_*$는 유일하다는 사실을 즉시 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem4&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 4.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $\lambda : S \to S'$ is a mapping of sets, there is a unique homomorphism $\bar{\lambda}$ making the following diagram commutative with the notation of the preceding theorem:&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;두 집합 사이의 사상 $\lambda : S \to S'$가 주어질 때, 위 정리에서의 노테이션 하에서 아래의 다이어그램이 가환이 되도록 하는 homomorphism $\bar{\lambda}$가 유일하게 존재한다.&lt;/div&gt;
$$ \begin{CD} S @&amp;gt;f_S&amp;gt;&amp;gt; \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \\ @V{\lambda}VV @VV\bar{\lambda}V \\ S' @&amp;gt;&amp;gt;f_{S'}&amp;gt; \mathbb{Z} \! \left&amp;lt; S' \right&amp;gt; \end{CD} $$&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let's define $\bar{\lambda} : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to \mathbb{Z} \! \left&amp;lt; S' \right&amp;gt;$ as&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ \bar{\lambda} : \sum_{x \in S} k_x \cdot x \mapsto \sum_{x \in S} k_x ( f_{S'} \circ \lambda )(x). $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Then it is obvious that $\bar{\lambda}$ is a homomorphism and that $f_{S'} \circ \lambda = \bar{\lambda} \circ f_S$. Thus, such homomorphism exists and now only showing uniqueness of such homomorphism is remaining.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let's focus on the condition that the diagram in the statement is commutative. We can rewrite the condition that the diagram in the statement is commutative to an equation $1 \cdot \lambda(x) = \bar{\lambda}(1 \cdot x)$. Suppose that a homomorphism $\underline{\lambda} : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to \mathbb{Z} \! \left&amp;lt; S' \right&amp;gt;$ also makes the diagram in the statement commutative. Then we can get $\underline{\lambda}(1 \cdot x) = 1 \cdot \lambda(x) = \bar{\lambda}(1 \cdot x)$ for every $x \in S$ and since $\{ 1 \cdot x \;|\; x \in S \}$ generates $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$ and $\bar{\lambda}$ and $\underline{\lambda}$ are homomorphisms, it is immediately verified that $\bar{\lambda} = \underline{\lambda}$. Therefore, the theorem is proved.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\bar{\lambda} : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to \mathbb{Z} \! \left&amp;lt; S' \right&amp;gt;$를 다음과 같이 정의하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$$ \bar{\lambda} : \sum_{x \in S} k_x \cdot x \mapsto \sum_{x \in S} k_x ( f_{S'} \circ \lambda )(x) $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러면 $\bar{\lambda}$가 homomorphism이고 $f_{S'} \circ \lambda = \bar{\lambda} \circ f_S$인 것은 자명하다. 따라서 조건을 만족하는 homomorphism은 존재하며, 이제 그러한 homomorphism이 유일하다는 것을 보이면 정리가 증명된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 위 다이어그램이 가환이라는 조건에 집중하자. 위 다이어그램이 가환이라는 조건은 $1 \cdot \lambda(x) = \bar{\lambda}(1 \cdot x)$라는 식으로 다시 쓸 수 있다. 이제 homomorphism $\underline{\lambda} : \mathbb{Z} \! \left&amp;lt; S \right&amp;gt; \to \mathbb{Z} \! \left&amp;lt; S' \right&amp;gt;$ 역시 조건을 만족한다고 가정하자. 그러면 임의의 $x \in S$에 대하여 $\underline{\lambda}(1 \cdot x) = 1 \cdot \lambda(x) = \bar{\lambda}(1 \cdot x)$가 성립함을 알 수 있으며, $\{ 1 \cdot x \;|\; x \in S \}$가 $\mathbb{Z} \! \left&amp;lt; S \right&amp;gt;$를 생성하고 $\bar{\lambda}$와 $\underline{\lambda}$가 homomorphism이므로 $\bar{\lambda} = \underline{\lambda}$인 것이 그 즉시 보여진다. 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem5&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 5.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Every abelian group $A$ is isomorphic to a factor group of a free abelian group $F$. If $A$ is finitely generated, show that one can select $F$ to be finitely generated also.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;임의의 아벨군 $A$에 대하여 $A$와 $F$의 몫군이 isomorphic하도록 하는 자유 아벨군 $F$가 항상 존재한다. 만약 $A$가 &lt;a href=&quot;https://chocobear.tistory.com/193#Definition6&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;유한생성군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이면, $F$ 역시 유한생성군이 되도록 선택할 수 있다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $S$ be a generating set of $A$ and choose a set $T$ and a surjective map $g : T \to S$. Now define a homomorphism $g_* : \mathbb{Z} \! \left&amp;lt; T \right&amp;gt; \to A$ as&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ g_* : \sum_{t \in T} k_t \cdot t \mapsto \sum_{t \in T} k_t g(t). $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $g$ is surjective and $S$ generates $A$, $g_*$ is also surjective. Therefore, $A \cong \mathbb{Z} \! \left&amp;lt; T \right&amp;gt; / \operatorname{ker} g_*$ by &lt;a href=&quot;https://chocobear.tistory.com/187#Part1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;the first isomorphism theorem for groups&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;. The second assertion is obvious since if $A$ is finitely generated, we can choose finite sets $S$ and $T$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$S$를 $A$의 생성 집합이라 하고, 전사함수 $g : T \to S$가 존재하도록 하는 집합 $T$를 생각하자. 이제 homomorphism $g_* : \mathbb{Z} \! \left&amp;lt; T \right&amp;gt; \to A$를 다음과 같이 정의하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$$ g_* : \sum_{t \in T} k_t \cdot t \mapsto \sum_{t \in T} k_t g(t) $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$g$가 전사함수이고 $S$가 $A$를 생성한다는 것으로부터 $g_*$ 역시 전사함수임을 알 수 있다. 따라서 &lt;a href=&quot;https://chocobear.tistory.com/187#Part1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;군에서의 제1동형 정리&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 의해 $A \cong \mathbb{Z} \! \left&amp;lt; T \right&amp;gt; / \operatorname{ker} g_*$임을 알 수 있다. 또한, $A$가 유한생성군이라면 유한집합인 $S$와 $T$를 고를 수 있으므로 $\mathbb{Z} \! \left&amp;lt; T \right&amp;gt;$ 역시 유한생성군이 된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Given an abelian group $A$ and a subgroup $B$, it is sometimes desirable to find a subgroup $C$ such that $A = B \oplus C$. The next lemma gives us a condition under which this is true.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;아벨군 $A$와 그의 부분군 $B$가 주어질 때, $A = B \oplus C$가 되도록 하는 부분군 $C$를 얻기 쉬울 때가 있다. 아래는 어떤 조건 아래에서 그러한지를 알려주는 보조정리이다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem6&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 6.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $f : A \to A'$ be a surjective homomorphism of abelian groups, and assume that $A'$ is free. Let $B$ be the kernel of $f$. Then there exists a subgroup $C$ of $A$ such that the restriction of $f$ to $C$ induces an isomorphism of $C$ with $A'$, and such that $A = B \oplus C$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;두 아벨군 사이의 homomorphism $f : A \to A'$가 전사함수이고 $A'$이 자유 아벨군이라고 하자. $B$를 $f$의 kernel이라고 하면 $f$의 $C$로의 축소가 $C$에서 $A'$으로 가는 isomorphism이 되도록 하면서 $A = B \oplus C$인 $A$의 부분군 $C$가 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\{ x'_i \}_{i \in I}$ be a basis of $A'$, and for each $i \in I$, let $x_i$ be an element of $A$ such that $f(x_i) = x'_i$. Let $C$ be the subgroup of $A$ generated by all elements $x_i$, $i \in I$. If we have a relation&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ \sum_{i \in I} n_i x_i = 0 $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;with integers $n_i$, which are equal to $0$ for all but a finite number of indices $i \in I$, then applying $f$ yeilds&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ 0 = \sum_{i \in I} n_i f(x_i) = \sum_{i \in I} n_i x'_i $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;hence all $n_i = 0$. So our family $\{ x_i \}_{i \in I}$ is a basis of $C$. Similarly, one sees that if $z \in C$ and $f(z) = 0$ then $z = 0$. Hence $B \cap C = \{ 0 \}$. Let $x \in A$. Since $f(x) \in A'$, there exists integers $n_i$, $i \in I$, such that&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;$$ f(x) = \sum_{i \in I} n_i x'_i. $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Applying $f$ to $x - \displaystyle \sum_{i \in I} n_i x_i$, we find that this element lies in the kernel of $f$, say $x - \displaystyle \sum_{i \in I} n_i x_i \in B$. From this we see $x \in B+C$, and hence finally that $A = B \oplus C$ is a direct sum, as contended.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\{ x'_i \}_{i \in I}$를 $A'$의 기저라고 하고 각 $i \in I$에 대하여 $x_i$를 $f(x_i) = x'_i$인 $x_i \in A$라고 하자. 이제 $C$를 $x_i$들로부터 생성된 $A$의 부분군이라고 하자. 만약 유한 개를 제외한 모든 것이 $0$인 정수 $n_i$들에 대하여 $\displaystyle \sum_{i \in I} n_i x_i = 0$이 성립한다면, $0 = \displaystyle \sum_{i \in I} n_i f(x_i) = \sum_{i \in I} n_i x'_i$임을 알 수 있으며, 따라서 모든 $n_i$는 $0$이다. 따라서 $\{ x_i \}_{i \in I}$는 $C$의 기저임을 알 수 있다. 비슷한 방법으로, $z \in C$이고 $f(z) = 0$이면 $z = 0$임을 알 수 있으며, 따라서 $B \cap C = \{ 0 \}$이다. 이제 $x \in A$라고 하자. $f(x) \in A'$이므로 $f(x) = \sum_{i \in I} n_i x'_i$가 성립하도록 하는 정수 $n_i$들이 존재한다. 함수 $f$를 $x - \displaystyle \sum_{i \in I} n_i x_i$에 적용하면, 이 원소가 $f$의 kernel의 원소가 됨을 알 수 있다. 즉, $x - \displaystyle \sum_{i \in I} n_i x_i \in B$이다. 이로부터 $x \in B+C$임을 알 수 있으며, 따라서 $A = B \oplus C$이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem7&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 7.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $A$ be a free abelian group, and let $B$ be a subgroup. Then $B$ is also a free abelian group.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;자유 아벨군 $A$와 그의 부분군 $B$가 주어졌다고 하자. 그러면 $B$ 역시 자유 아벨군이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $X$ be a set of every subset $S$ of $B$ such that $S$ is a basis of $\left&amp;lt; S \right&amp;gt;$. Since $A$ is free abelian, there exists an element $b \in B$ such that $\left&amp;lt; b \right&amp;gt; \cong \mathbb{Z}$ so $X$ is nonempty.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let $T \in X$ be a maximal element of $X$. Then $T$ is a basis of $\left&amp;lt; T \right&amp;gt;$. If $B / \left&amp;lt; T \right&amp;gt;$ is a trivial group, then $B = \left&amp;lt; T \right&amp;gt;$ so $B$ is a free abelian group. Now suppose that $B / \left&amp;lt; T \right&amp;gt;$ is not trivial. Then there exists an element $x \in B$ such that $x \notin \left&amp;lt; T \right&amp;gt;$ and $\{ m \in \mathbb{Z} \;|\; mx \in \left&amp;lt; T \right&amp;gt; \}$ is either $\{ 0 \}$ or $n\mathbb{Z}$ for some integer $n&amp;gt;1$. If $\{ m \in \mathbb{Z} \;|\; mx = \left&amp;lt; T \right&amp;gt; \} = \{ 0 \}$, then $\{ x \} \cup T$ is a basis of $\left&amp;lt; x \right&amp;gt; + \left&amp;lt; T \right&amp;gt;$, obviously, and this contradicts the maximality of $T$. Thus, $\{ m \in \mathbb{Z} \;|\; mx = \left&amp;lt; T \right&amp;gt; \}$ must be $n\mathbb{Z}$ for some integer $n &amp;gt; 1$. This means that $nx = \displaystyle \sum_{t \in T} m_t t$ for some integers $m_t$, $t \in T$ such that $m_t = 0$ for all but a finite number of elements $t \in T$. So we can take an elemtn $t_0 \in T$ such that $m_{t_0} \neq 0$ since $nx \neq 0$ (because $A$ is free abelian). It is then clear that $\left&amp;lt; nx \right&amp;gt; + \left&amp;lt; T \setminus \{ t_0 \} \right&amp;gt; = \left&amp;lt; T \right&amp;gt;$. Thus makes $\{ x \} \cup \left( T \setminus \{ t_0 \} \right)$ be a maximal element of $X$. So by the &lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/70&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;well-ordering theorem&lt;/b&gt;&lt;/a&gt;&lt;/span&gt; and the &lt;a href=&quot;https://chocobear.tistory.com/63&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;transfinite induction&lt;/span&gt;&lt;/b&gt;&lt;/a&gt;, we can construct a maximal element $T'$ of $X$ which generates $B$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Hence, if a maximal element of $X$ exists, $B$ is a free abelian group, clearly. By the &lt;a href=&quot;https://chocobear.tistory.com/69&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Zorn's lemma&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;, if every chain in $X$ has an upper bound in $X$, $X$ always has a maximal element. So if we show that $\bigcup \mathcal{C} \in X$ for any chains $\mathcal{C}$ in $X$, the theorem will be proved.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let $\mathcal{C}$ be a chain in $X$. Fix an element $x \in \left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$. Then $x$ has an expression as a linear combination $x = \displaystyle \sum_{i \in \bigcup \mathcal{C}} m_i i$ with $m_i \in \mathbb{Z}$ and $m_i = 0$ for all but a finite number of elements $i \in \bigcup \mathcal{C}$. Now $x = \displaystyle \sum_{i \in \bigcup \mathcal{C}} m_i i = \sum_{i \in \bigcup \mathcal{C}} m'_i i$ with $m_i, m'_i \in \mathbb{Z}$ and $m_i = m'_i = 0$ for all but a finite number of elements $i \in \bigcup \mathcal{C}$. Now define a set $S = \{ i \in \bigcup \mathcal{C} \;|\; m_i \neq 0 \lor m'_i \neq 0 \}$ and $\mathcal{D}$ a subset of $\mathcal{C}$ such that each element of $\mathcal{D}$ has $S$ as a subset. Then $\mathcal{D}$ has a minimum element $M \in \mathcal{D}$ since $\mathcal{C}$ is a chain in $X$. Since $M$ is an element of $X$, $M$ is a basis of $\left&amp;lt; M \right&amp;gt;$ and, hence, $m_i = m'_i$ for every $i \in \bigcup \mathcal{C}$ by the fact that $S \subseteq M$ and the definition of $S$. Hence, $x$ has a unique expression as a linear combination. Since the choice of $x \in \left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$ is arbitrary, $\bigcup \mathcal{C}$ is a basis of $\left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$ and $\bigcup \mathcal{C}$ is therefore an element of $X$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Therefore, every chain in $X$ has an upper bound so $B$ is a free abelian group.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now we shall prove the second assertion. Let $\mathcal{A}$ be a basis of $A$ and $\mathcal{B}$ be a basis of $B$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;집합 $X$를 $S$가 $\left&amp;lt; S \right&amp;gt;$의 기저가 되도록 하는 $B$의 부분집합 $S$들의 집합으로 정의하자. $A$가 자유 아벨군이므로 $\left&amp;lt; b \right&amp;gt; \cong \mathbb{Z}$이도록 하는 $b \in B$가 존재하며, 따라서 $X$는 공집합이 아니다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $T \in X$가 $X$의 극대원소라고 가정하자. 그러면 $T$는 $\left&amp;lt; T \right&amp;gt;$의 기저이다. 만약 $B / \left&amp;lt; T \right&amp;gt;$가 자명군이라면, $B = \left&amp;lt; T \right&amp;gt;$이므로 $T$가 $B$의 기저가 되어 $B$가 자유 아벨군이 된다. 이번엔 $B / \left&amp;lt; T \right&amp;gt;$가 자명군이 아니라고 가정하자. 그러면 $x \notin \left&amp;lt; T \right&amp;gt;$인 $x \in B$가 존재하며, $\{ m \in \mathbb{Z} \;|\; mx \in \left&amp;lt; T \right&amp;gt; \}$는 $\{ 0 \}$이거나 어떤 정수 $n &amp;gt; 1$에 대하여 $n \mathbb{Z}$이다. 만약 $\{ m \in \mathbb{Z} \;|\; mx \in \left&amp;lt; T \right&amp;gt; \} = \{ 0 \}$이라면, $\{ x \} \cup T$가 $\left&amp;lt; x \right&amp;gt; + \left&amp;lt; T \right&amp;gt;$의 기저가 되므로 $T$가 $X$의 극대원소라는 사실에 모순이 발생한다. 따라서 $\{ m \in \mathbb{Z} \;|\; mx \in \left&amp;lt; T \right&amp;gt; \}$는 어떤 정수 $n &amp;gt; 1$에 대하여 $n\mathbb{Z}$이다. 이는 곧 $nx = \displaystyle \sum_{t \in T} m_t t$이고 유한 개의 $t$를 제외한 모든 $t \in T$에 대하여 $m_t = 0$이도록 하는 정수들 $m_t$가 존재함을 의미한다. 이때 $x \neq 0$이므로 $m_{t_0} \neq 0$이도록 하는 $t_0 \in T$가 존재하며, $\left&amp;lt; nx \right&amp;gt; + \left&amp;lt; T \setminus \{ t_0 \} \right&amp;gt; = \left&amp;lt; T \right&amp;gt;$임이 자명하다. 따라서 $\{ x \} \cup \left( T \setminus \{ t_0 \} \right)$가 $X$의 극대원소가 된다. 그러므로 &lt;a href=&quot;https://chocobear.tistory.com/70&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;정렬 정리&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;와 &lt;a href=&quot;https://chocobear.tistory.com/63&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;초한 귀납법&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 의해 $B$를 생성하는 $X$의 극대원소 $T'$을 구성할 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 $X$의 극대원소가 존재하면 $B$가 자유 아벨군임을 알 수 있다. 이때, &lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/69&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;초른 보조정리&lt;/a&gt;&lt;/b&gt;&lt;/span&gt;에 의해 $X$의 모든 사슬이 상계를 가지면 $X$의 극대원소가 존재하므로 $X$의 임의의 사슬 $\mathcal{C}$에 대하여 $\bigcup \mathcal{C} \in X$가 성립한다면 정리가 증명된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $\mathcal{C}$를 $X$의 사슬이라고 하고 원소 $x \in \left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$를 고정하자. 그러면 $x$는 유한 개의 $i$를 제외한 모든 $i \in \bigcup \mathcal{C}$에 대하여 $m_i = 0$이고 $x = \displaystyle \sum_{i \in \bigcup \mathcal{C}} m_i i$가 성립하도록 하는 정수 $m_i \in \mathbb{Z}$가 존재한다. 이제 $x$의 두 선형결합 표현 $x = \displaystyle \sum_{i \in \bigcup \mathcal{C}} m_i i = \sum_{i \in \bigcup \mathcal{C}} m'_i i$를 생각하고 집합 $S$를 $S = \{ i \in \bigcup \mathcal{C} \;|\; m_i \neq 0 \lor m'_i \neq 0 \}$와 같이 정의하고 $S$를 부분집합으로 가지는 집합들로만 이루어진 $\mathcal{C}$의 부분집합을 $\mathcal{D}$라고 하자. 그러면 $\mathcal{C}$가 사슬이라는 사실로부터 $\mathcal{D}$는 최소원소 $M$을 가진다는 것을 알 수 있으며, $M \in X$이므로 $M$은 $\left&amp;lt; M \right&amp;gt;$의 기저임을 알 수 있다. 따라서 $S \subseteq M$이라는 사실과 $S$의 정의로부터 임의의 $i \in \bigcup \mathcal{C}$에 대하여 $m_i = m'_i$가 성립함을 알 수 있다. 따라서 $x$의 선형결합 표현은 유일하게 존재하며, $x \in \left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$의 선택이 임의적이므로 $\bigcup \mathcal{C}$는 $\left&amp;lt; \bigcup \mathcal{C} \right&amp;gt;$의 기저이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 $X$의 모든 사슬은 상계를 가지며, 이로 인해 $B$는 자유 아벨군이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem8&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 8.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a free abelian group. Any two bases of $G$ have the same cardinality.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;자유 아벨군 $G$가 주어졌다고 하자. $G$의 임의의 두 기저는 같은 기저를 가진다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;If $G$ has a finite basis $X = \{ x_1, x_2, \cdots, x_n \}$, then $G \cong \displaystyle \bigoplus_{i=1}^n \mathbb{Z}$ so $G / 2G \cong \left( \displaystyle \bigoplus_{i=1}^n \mathbb{Z} \right) / \left( \bigoplus_{i=1}^n 2\mathbb{Z} \right) \cong \bigoplus_{i=1}^n \left( \mathbb{Z}/2\mathbb{Z} \right)$ and thus $\lvert G/2G \rvert = 2^n = 2^{\lvert X \rvert}$. Hence every finite basis of $G$ has the same cardinality. Now show that if $G$ has a finite basis, then there is no infinite basis of $G$. Let $X = \{ x_1, x_2, \cdots, x_n \}$ and $Y$ be a basis of $G$ and a finite set $\{ y_1, y_2, \cdots, y_s \}$ be a subset of $Y$. Let $H = \left&amp;lt; y_1, y_2, \cdots, y_s \right&amp;gt;$ and $K = \left&amp;lt; Y \setminus \{ y_1, y_2, \cdots, y_s \} \right&amp;gt;$ then it is clear that $G = H \oplus K$ so $G/2G = \left( H \oplus K \right) / \left( 2H \oplus 2K \right) \cong (H/2H) \oplus (K/2K)$. Thus makes $H/2H$ be a subgroup of $G/2G$ and, hence, $2^s = \lvert H/2H \rvert \leq \lvert G/2G \rvert = 2^n$. so it is obvious that $s \leq n$. Therefore, $Y$ is not an infinite set so if $G$ has a finite basis, then every basis of $G$ has the same cardinality.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now suppose that $G$ has an infinite basis. Let $B_1$ and $B_2$ be two bases of $G$. Then $B_1$ and $B_2$ are infinite by the fact that there is no infinite basis of a free abelian group which has a finite basis as shown above. Now let $\mathcal{P}_\mathrm{fin} (B_1)$ be the set of the finite sets of $B_1$ and a map $f : B_2 \to \mathcal{P}_\mathrm{fin} (B_1)$ send an element $b \in B_2$ to the smallest set $U \in \mathcal{P}_\mathrm{fin}(B_1)$ such that $b \in \left&amp;lt; U \right&amp;gt;$. Then since every finite basis of a free abelian group has the same cardinality, $f^{-1}(U)$ is finite for every $U \in \mathcal{P}_\mathrm{fin} (B_1)$ thus makes $\lvert B_2 \rvert \leq \aleph_0 \times \lvert \mathcal{P}_\mathrm{fin}(B_1) \rvert = \aleph_0 \times \lvert B_1 \rvert = \lvert B_1 \rvert$. In the same way, we can show that $\lvert B_1 \rvert \leq \lvert B_2 \rvert$. By the &lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/24&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;Cantor-Bernstein theorem&lt;/a&gt;&lt;/b&gt;&lt;/span&gt;, $\lvert B_1 \rvert = \lvert B_2 \rvert$ and therefore the theorem is proved.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;만약 $G$가 유한 기저 $X = \{ x_1, x_2, \cdots, x_n \}$를 가진다면, $G \cong \displaystyle \bigoplus_{i \in I}^n \mathbb{Z}$이며, 이로 인해 $G/2G \cong \left( \displaystyle \bigoplus_{i \in I}^n \mathbb{Z} \right) / \left( \bigoplus_{i \in I}^n 2\mathbb{Z} \right) \cong \bigoplus_{i \in I}^n \left( \mathbb{Z}/2\mathbb{Z} \right)$임을 알 수 있고, 따라서 $\lvert G/2G \rvert = 2^n = 2^{\lvert X \rvert}$임을 알 수 있다. 따라서 $G$의 모든 유한 기저의 기수는 같다. 이제 $G$가 유한 기저를 가지면, $G$의 무한 기저가 존재하지 않는다는 것을 보이자. 유한집합 $X = \{ x_1, x_2, \cdots, x_n \}$와 집합 $Y$가 $G$의 기저이고 유한집합 $\{ y_1, y_2, \cdots, y_s \}$가 $Y$의 부분집합이라고 하자. 이제 $H = \left&amp;lt; y_1, y_2, \cdots, y_s \right&amp;gt;$이고 $K = \left&amp;lt; Y \setminus \{ y_1, y_2, \cdots, y_s \} \right&amp;gt;$라고 하면 $G = H \oplus K$임은 자명하다. 따라서 $G/2G = \left( H \oplus K \right) / \left( 2H \oplus 2K \right) \cong (H/2H) \oplus (K/2K)$임을 알 수 있다. 따라서 $H/2H$가 $G/2G$의 부분군이 되며, 이로 인해 $2^s = \lvert H/2H \rvert \leq \lvert G/2G \rvert 2^n$을 얻을 수 있다. 즉, $s \leq n$이 성립한다. 따라서 $Y$는 무한집합이 될 수 없으며, 따라서 $G$의 무한 기저는 존재하지 않는다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번엔 $G$가 무한 기저를 가진다고 가정하자. 그러면 위에서 보인 사실에 의해 $G$의 두 기저 $B_1$과 $B_2$는 무한집합이다. 이제 $\mathcal{P}_\mathrm{fin} (B_1)$을 $B_1$의 모든 유한 부분집합의 집합으로 정의하고 사상 $f : B_2 \to \mathcal{P}_\mathrm{fin} (B_1)$를 $b \in B_2$를 $b \in \left&amp;lt; U \right&amp;gt;$가 성립하도록 하는 가장 작은 집합 $U \in \mathcal{P}_\mathrm{fin} (B_1)$로 대응되는 사상으로 정의하자. 그러면 자유 아벨군의 모든 유한 기저가 같은 기수를 가진다는 사실에 의해 임의의 $U \in \mathcal{P}_\mathrm{fin} (B_1)$에 대하여 $f^{-1} (U)$가 유한집합임을 알 수 있다. 따라서 $\lvert B_2 \rvert \leq \aleph_0 \times \lvert \mathcal{P}_\mathrm{fin} (B_1) \rvert = \aleph_0 \times \lvert B_1 \rvert = \lvert B_1 \rvert$가 성립하며, 같은 방법으로 $\lvert B_1 \rvert \leq \lvert B_2 \rvert$임을 보일 수 있다. 따라서 &lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/24&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;칸토어-베른슈타인 정리&lt;/a&gt;&lt;/b&gt;&lt;/span&gt;에 의해 $\lvert B_1 \rvert = \lvert B_2 \rvert$이며, 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The theorem above gives us the concept like dimension of a free abelian group. The number of elements in a basis of a free abelian group $A$ will be called the &lt;b&gt;rank&lt;/b&gt; of $A$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Moreover, free abelian groups have many other properties, but the rest is omitted for concison and if the properties not introduced in this post are used in other posts to be written later, we will prove them in those posts.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위의 정리는 우리에게 자유 아벨군에 일종의 차원과 비슷한 개념을 부여할 수 있음을 알려준다. 자유 아벨군의 기저의 원소의 개수를 우리는 자유 아벨군의 &lt;b&gt;계수 ( Rank )&lt;/b&gt;라고 부른다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이 외에도 자유 아벨군은 많은 성질들을 가지지만, 분량상 나머지는 생략했으며, 이후 작성하게 될 글들에서 이번 글에서 소개하지 않은 성질을 사용할 경우, 해당 글에서 증명하도록 하겠다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Free Abelian Group</category>
      <category>자유 아벨군</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/210</guid>
      <comments>https://chocobear.tistory.com/210#entry210comment</comments>
      <pubDate>Sat, 26 Feb 2022 08:58:44 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물다섯 번째 이야기  |  여러 가지 Solvable Group ( Various Examples of Solvable Groups )</title>
      <link>https://chocobear.tistory.com/206</link>
      <description>&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 &lt;a href=&quot;https://chocobear.tistory.com/194#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Solvable Group&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;의 여러 예시들을 소개하려 한다. 또한, 이후에 포스팅하게 될 여러 글들에서 소개하는 개념을 이용하여 solvable하다는 것을 보일 수 있는 예시들이 추가될 때마다 이 글에 해당 내용을 추가할 예정이다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;This post will introduce you various examples of &lt;a href=&quot;https://chocobear.tistory.com/194#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;solvable groups&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;. Also, the contents of this post will be added whenever the examples are added by posts that will be posted later.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;h3 data-ke-size=&quot;size23&quot;&gt;&lt;a name=&quot;toc&quot;&gt;&lt;/a&gt;&lt;b&gt;Table of Contents&lt;/b&gt;&lt;/h3&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Example1&quot;&gt;Example 1. Abelian Groups&lt;/a&gt;&lt;/li&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Example2&quot;&gt;Example 2. Finite $p$-Groups&lt;/a&gt;&lt;/li&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Example3&quot;&gt;Example 3. The Groups Whose Order is A Product of Two Prime Numbers&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/ul&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Example1&quot;&gt;&lt;/a&gt;&lt;a href=&quot;#toc&quot;&gt;&lt;b&gt;Example 1.&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;solvable group의 가장 기본적인 예시는 아벨군이다. 아벨군은 모든 부분군이 &lt;a href=&quot;https://chocobear.tistory.com/128#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;정규부분군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이므로 자명부분군 $\mathbf{1}$ 역시 정규부분군으로 가진다. 따라서 아벨군은 solvable하다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The basic example of solvable groups are abelian groups. They are trivially solvable since a &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/194#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;normal series&lt;/b&gt;&lt;/a&gt; being given by just the group itself and the trivial group.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Example2&quot;&gt;&lt;/a&gt;&lt;a href=&quot;#toc&quot;&gt;&lt;b&gt;Example 2.&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;모든 유한 &lt;a href=&quot;https://chocobear.tistory.com/204#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;$p$-군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;은 solvable하다. 아래의 정리를 보라.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Every finite &lt;a href=&quot;https://chocobear.tistory.com/204#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;$p$-group&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; is solvable. Look at the theorem below.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem2&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a finite $p$-group. Then $G$ is solvable. If its order is greater than $1$, then $G$ has a non-trivial &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/198#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;center&lt;/b&gt;&lt;/a&gt;.&lt;/div&gt;
&lt;div&gt;$G$가 유한 $p$-군이라고 하자. 그러면 $G$는 solvable하다. 만약 $G$의 위수가 $1$보다 크다면, $G$는 비자명 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/198#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;Center&lt;/b&gt;&lt;/a&gt;를 가진다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The first assertion follows from the second, since if $G$ has center $Z$, and we have an &lt;a href=&quot;https://chocobear.tistory.com/194#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;abelian tower&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; for $G/Z$ by induction, we can lift this abelian tower to $G$ to show that $G$ is solvable. To prove the second assertion, we use the class equation $$ \left( G : \mathbf{1} \right) = \left( Z : \mathbf{1} \right) + \sum \left( G : G_x \right), $$ the sum being taken over certain $x$ for which $\left( G : G_x \right) \neq 1$. Then $p$ divides $\left( G : \mathbf{1} \right)$ and also divides every term in the sum, so that $p$ divides the order of the center, as was to be shown.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;만약 $G$가 비자명 center $Z$를 가진다면 귀납법에 의해 $G/Z$의 &lt;a href=&quot;https://chocobear.tistory.com/194#Definition1&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;abelian tower&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;를 얻을 수 있고, 이를 이용하여 $G$의 abelian tower를 구성할 수 있으므로 $G$가 solvable하다는 것이 보여진다. 따라서 $G$가 비자명 center $Z$를 가진다는 것만 보여도 충분하다. 이를 보이기 위해 우리는 class equation $$ \left( G : \mathbf{1} \right) = \left( Z : \mathbf{1} \right) + \sum \left( G : G_x \right) $$를 사용할 것이다. 이때, 우변의 시그마는 $\left( G : G_x \right) \neq 1$인 $x$에 대하여 $\left( G : G_x \right)$의 합을 취한 것이다. 그러면 $p$가 $\left( G : \mathbf{1} \right)$를 나누고 시그마의 각 항을 나누므로 $\left( Z : \mathbf{1} \right)$ 역시 $p$로 나누어 떨어짐을 알 수 있으며, 따라서 $G$는 비자명 center $Z$를 가진다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Corollary 2.1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a finite $p$-group which is not of order $1$. Then there exists a sequence of subgroups $$ \mathbf{1} = G_0 \subseteq G_1 \subseteq \cdots \subseteq G_n = G $$ such that $G_i$ is normal in $G$ and $G_{i+1} / G_i$ is cyclic of order $p$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$G$를 위수가 $1$이 아닌 유한 $p$-군이라고 하자. 그러면 각 $G_i$가 $G$의 정규부분군이고 각 $G_{i+1} / G_i$가 위수가 $p$인 순환군이도록 하는 부분군열 $$ \mathbf{1} = G_0 \subseteq G_1 \subseteq \cdots \subseteq G_n = G $$가 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $G$ has a non-trivial center, there exists an element $a \neq e$ in the center of $G$, and such that $a$ has order $p$. Let $H$ be the cyclic group generated by $a$. By induction, if $G \neq H$, we can find a sequence of subgroups as stated above in the factor group $G/H$. Taking the inverse image of this tower in $G$ gives us the desired sequence in $G$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$가 비자명 center를 가지므로 $G$의 center는 위수가 $p$인 항등원이 아닌 원소 $a$를 가진다. $H$를 $a$로부터 생성된 순환군이라고 하자. 만약 $G = H$라면 따름정리가 성립함은 자명하며, $G \neq H$라면, $G/H$ 역시 유한 $p$-군이므로 귀납법을 적용할 수 있다. $G/H$가 조건을 만족하는 부분군열을 가진다고 하면 이로부터 자연스럽게 $G$의 부분군열 역시 구성 가능하다. 따라서 귀납법에 의해 따름정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Example3&quot;&gt;&lt;/a&gt;&lt;a href=&quot;#toc&quot;&gt;&lt;b&gt;Example 3.&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위수가 두 소수의 곱인 군은 solvable하다. 아래를 보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Every group whose order is the product of two primes is solvable. Look at below.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;#Lemma3&quot;&gt;&lt;/a&gt;&lt;b&gt;Lemma 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a finite group and let $p$ be the smallest prime dividing the order of $G$. Let $H$ be a subgroup of index $p$. Then $H$ is normal.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;유한군 $G$가 주어졌다고 하고 $p$가 $G$의 위수의 가장 작은 소인수라고 하자. 그러면 $\left( G : H \right) = p$인 $G$의 부분군 $H$는 언제나 $G$의 정규부분군이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $N_G(H) = N$ be the normalizer of $H$. Then $N = G$ or $N = H$. If $N = G$, we are done. Suppose $N = H$. Then the orbit of $H$ under conjugation has $p = \left( G : H \right)$ elements, and the representation of $G$ on this orbit gives a homomorphism of $G$ into the symmetric group on $p$ elements, whose order is $p!$ since each element of $G$ one-to-one corresponds the orbit of $H$ to itself. Let $K$ be the kernel of that homomorphism. Then $K$ is the intersection of &lt;a href=&quot;https://chocobear.tistory.com/202#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;stabilizer subgroups&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;, and the stabilizer subgroup of $H$ is $H$ by assumption, so $K \subseteq H$. If $K \neq H$, then from $$ \left( G : K \right) = \left( G : H \right) \left( H : K \right) = p \left( H : K \right), $$ and the fact that only the first power of $p$ divides $p!$, we conclude that some prime dividing $(p-1)!$ also divides $\left( H : K \right)$, which contradicts the assumption that $p$ is the smallest prime dividing the order of $G$. Thus $K = H$ and this means that $H$ is a subgroup of the stabilizer subgroup of every conjugation of $H$. So $hgHg^{-1}h^{-1} = gHg^{-1}$ for every $g \in G,\;h \in H$ and this implies that $g^{-1}hg \in H$ for every $g \in G,\;h \in H$ by &lt;a href=&quot;https://chocobear.tistory.com/198#Theorem1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;the fact&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; that the normalizer of $H$ is the largest subgroup of $G$ in which $H$ is normal. And this means that $g^{-1}Hg \subseteq H$ for every $g \in G$ and this contradicts the assumption that $N= H$ i.e., $H$ is not normal in $G$. Therefore, we can get $N = G$ and the proof is complete.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$N_G(H) = N$을 $H$의 normalizer라고 하자. 그러면 $N = G$이거나 $N = H$이다. 만약 $N = G$라면 보조정리가 성립함은 자명하다. 따라서 $N = H$라고 가정하자. 그러면 켤레연산 아래에서 $H$의 궤도는 $p = \left( G : H \right)$개의 원소를 가지며, $G$의 각 원소는 켤레로서 작용하면서 $H$의 궤도에서 $H$의 궤도로 가는 일대일 대응을 만들어내므로 켤레연산은 $G$로부터 $S_p$로 가는 homomorphism을 구성해낸다. $K$를 이 homomorphism의 kernel이라고 하면 $K$는 모든 &lt;b&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/202#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;안정자군&lt;/a&gt;&lt;/span&gt;&lt;/b&gt;의 교집합임을 알 수 있다. 이때, 가정에 의해 $H$의 안정자군이 $H$이므로 $K \subseteq H$임을 알 수 있다. 만약 $K \neq H$라면, $$ \left( G : K \right) = \left( G : H \right) \left( H : K \right) = p \left( H : K \right) $$라는 사실과 $p!$을 나누는 $p$의 가장 큰 거듭제곱이 $p$라는 사실로부터 $(p-1)!$의 어떤 소인수가 $\left( H : K \right)$ 역시 나눈다는 것을 알 수 있으며, 이는 $p$가 $G$의 위수의 가장 작은 소인수라는 가정에 모순된다. 따라서 $K = H$라는 것을 알 수 있으며, 이는 곧 $H$가 $H$의 모든 켤레의 안정자군의 부분군이라는 것을 의미한다. 따라서 임의의 $g \in G,\;h \in H$에 대하여 $hgHg^{-1}h^{-1} = gHg^{-1}$이 성립하며, &lt;a href=&quot;https://chocobear.tistory.com/198#Theorem1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;이 정리&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에&amp;nbsp;의해 임의의 $g \in G,\;h \in H$에 대하여 $g^{-1}hg \in H$임을 알 수 있다. 이는 곧 임의의 $g \in G$에 대하여 $g^{-1}Hg \subseteq H$가 성립함을 의미하며, 이는 $N = H$ 즉, $H$가 $G$의 정규부분군이 아니라는 가정에 모순된다. 따라서 $N = G$임을 얻으며, 보조정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $p$ and $q$ are primes and let $G$ be a group of order $pq$. Then $G$ is solvable.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$p,\;q$를 소수라 하고 $G$를 위수가 $pq$인 군이라고 하자. 그러면 $G$는 solvable하다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;If $p = q$, then it is obvious that $G$ is solvable by the &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;. Suppose $p &amp;lt; q$. Let $Q$ be a &lt;a href=&quot;https://chocobear.tistory.com/204#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Sylow subgroup&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of order $q$. Then $Q$ has index $p$, so by the lemma, $Q$ is normal and the factor group has order $p$. But a group of prime order is cyclic, whence the theorem follows.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$p = q$인 경우, $G$가 $p$-군이므로 &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;에 의해 $G$가 solvable함이 자명하다. 이제 $p &amp;lt; q$라고 가정하자. $Q$를 위수 $q$인 &lt;a href=&quot;https://chocobear.tistory.com/204#Definition1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;쉴로브 부분군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이라고 하자. 그러면 $\left( G : Q \right) = p$이므로 보조정리에 의해 $Q$가 $G$의 정규부분군이며, $G/Q$의 위수가 $p$임을 알 수 있다. 위수가 소수인 군은 언제나 순환군이라는 사실로부터, 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Solvable Group</category>
      <category>가해군</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/206</guid>
      <comments>https://chocobear.tistory.com/206#entry206comment</comments>
      <pubDate>Mon, 21 Feb 2022 08:22:47 +0900</pubDate>
    </item>
    <item>
      <title>카테고리별 참고문헌 목록 | Reference List by Category</title>
      <link>https://chocobear.tistory.com/notice/205</link>
      <description>&lt;p data-ke-size=&quot;size16&quot;&gt;본 공지는 각 카테고리별로 어떤 서적을 참고하였는지를 나타내는 목록이다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;h3 data-ke-size=&quot;size23&quot;&gt;&lt;b&gt;Reference List by Category&lt;/b&gt;&lt;/h3&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size18&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;수학 | Mathematics&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/논리학%20| Mathematical Logic&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;논리학 | Mathematical Logic&lt;/b&gt;&lt;/a&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;Enderton, H. B. (2001). &lt;i&gt;A mathematical introduction to logic&lt;/i&gt;. Elsevier.&lt;/li&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Mendelson, E. (2009).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Introduction to mathematical logic&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;. Chapman and Hall/CRC.&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Cameron, P. J. (2012).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Sets, logic and categories&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;. Springer Science &amp;amp; Business Media.&lt;/span&gt;&lt;/span&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/정수론%20| Number Theory&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;정수론 | Number Theory&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/집합론%20| Set Theory&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;집합론 | Set Theory&lt;/b&gt;&lt;/a&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Hrbacek, K., &amp;amp; Jech, T. (1999).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Introduction to set theory, revised and expanded&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;. Crc Press.&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Jech, T. J., Jech, T., Jech, T. J., Mathematician, G. B., Jech, T. J., &amp;amp; Math&amp;eacute;maticien, G. B. (2003).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Set theory&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span&gt;&amp;nbsp;&lt;/span&gt;(Vol. 14). Berlin: Springer.&lt;/span&gt;&lt;/span&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/선형대수학%20| Linear Algebra&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;선형대수학 | Linear Algebra&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/추상대수학%20| Abstract Algebra&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;추상대수학 | Abstract Algebra&lt;/b&gt;&lt;/a&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Lang, S. (2012).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Algebra&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span&gt;&amp;nbsp;&lt;/span&gt;(Vol. 211). Springer Science &amp;amp; Business Media.&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Hungerford, T. W. (2012).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Abstract algebra: an introduction&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;. Cengage Learning.&lt;/span&gt;&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Fraleigh, J. B. (2003).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;A first course in abstract algebra&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;. Pearson Education India.&lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/해석학%20| Analysis&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;해석학 | Analysis&lt;/b&gt;&lt;/a&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;Rudin, W. (1976).&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/span&gt;&lt;i&gt;Principles of mathematical analysis&lt;/i&gt;&lt;span style=&quot;background-color: #ffffff; color: #222222;&quot;&gt;&lt;span&gt;&amp;nbsp;&lt;/span&gt;(Vol. 3). New York: McGraw-hill.&lt;/span&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/수학/기타%20| Uncategorized&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;기타 | Uncategorized&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size18&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;컴퓨터과학 | Computer Science &amp;amp; Engineering&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/그래프%20이론 | Graph Theory&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;그래프 이론 | Graph Theory&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/알고리즘%20| Algorithms&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;알고리즘 | Algorithms&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/자료구조%20| Data Structures&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;자료구조 | Data Structures&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/문제풀이%20| Problem Solving&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;문제풀이 | Problem Solving&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/CPU%20설계 | CPU Architecture&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;CPU 설계 | CPU Architecture&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/리눅스%20서버 | Linux Server&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;리눅스 서버 | Linux Server&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/컴퓨터과학/기타%20| Uncategorized&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;기타 | Uncategorized&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size18&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;물리 | Physics&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리/고전역학%20| Classical Mechanism&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;고전역학 | Classical Mechanism&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리/수리물리%20| Mathematical Physics&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;수리물리 | Mathematical Physics&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리/전자기학%20| Electromagnetism&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;전자기학 | Electromagnetism&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리/문제풀이%20| Problem Solving&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;문제풀이 | Problem Solving&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/물리/기타%20| Uncategorized&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;기타 | Uncategorized&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size18&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;화학 | Chemistry&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학/물리화학%20| Physical Chemistry&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;물리화학 | Physical Chemistry&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학/분석화학%20| Analytical Chemistry&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;분석화학 | Analytical Chemistry&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학/유기화학%20| Organic Chemistry&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;유기화학 | Organic Chemistry&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학/문제풀이%20| Problem Solving&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;문제풀이 | Problem Solving&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/화학/기타%20| Uncategorized&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;기타 | Uncategorized&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size18&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/생명과학&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;생명과학 | Biology&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/생명과학/생명과학%20| Biology&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;생명과학 | Biology&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/생명과학/진화생물학%20| Evolutionary Biology&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;진화생물학 | Evolutionary Biology&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/생명과학/세포생물학%20| Cell Biology&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;세포생물학 | Cell Biology&lt;/b&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/category/기타&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;기타 | Uncategorized&lt;/b&gt;&lt;/a&gt;&lt;/p&gt;</description>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/notice/205</guid>
      <pubDate>Mon, 21 Feb 2022 07:22:31 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물네 번째 이야기  |  쉴로브 정리 ( Sylow Theorems )</title>
      <link>https://chocobear.tistory.com/204</link>
      <description>&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, I would like to talk about Sylow theorems. The Sylow theorems are a collection of theorems named after the Norwegian mathematician &lt;a href=&quot;https://en.wikipedia.org/wiki/Peter_Ludwig_Mejdell_Sylow&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;Sylow&lt;/span&gt;&lt;/b&gt;&lt;/a&gt; that give detailed information about the number of subgroups of fixed order that a given finite group contains. The Sylow theorems form a fundamental part of finite group theory and have very important applications in the classification of finite simple groups. Before we get to know about Sylow theorems, let's find out what Sylow subgroups are.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 쉴로브 정리에 대해 알아보려 한다. 쉴로브 정리는 노르웨이 수학자 &lt;a href=&quot;https://en.wikipedia.org/wiki/Peter_Ludwig_Mejdell_Sylow&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;쉴로브&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;가 제시한 세 개의 정리로, 어떤 유한군의 부분군 중 특정 위수를 가진 부분군의 개수에 대한 정보를 알려주는 정리들이다. 쉴로브 정리는 유한군 이론의 가장 기본적인 정리 중 하나이며, 유한단순군의 분류작업에 사용된 중요한 정리 중 하나이다. 쉴로브 정리에 대해 알아보기에 앞서, 쉴로브 부분군이 무엇인지부터 알아보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. The $p$-Groups &amp;amp; The Sylow $p$-Subgroups&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. $p$-군과 쉴로브 $p$-부분군 ( $p$-Group &amp;amp; Sylow $p$-Subgroup )&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $p$ be a prime number. By a &lt;b&gt;$p$-Group&lt;/b&gt;, we mean a finite group whose order is a power of $p$ i.e., a finite group whose order is $p^n$ for some integer $n \geq 0$. Let $G$ be a finite group and $H$ a subgroup. We call $H$ a &lt;b&gt;$p$-Subgroup&lt;/b&gt; of $G$ if $H$ is a $p$-group. We call $H$ a &lt;b&gt;$p$-Sylow Subgroup&lt;/b&gt; if $H$ is a maximal $p$-subgroup of $G$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;소수 $p$가 주어졌다고 하자. 이때, &lt;b&gt;$p$-군 ( $p$-Group )&lt;/b&gt;이란, 위수가 $p$의 거듭제곱인 군을 말한다. 즉, 위수가 어떤 정수 $n \geq 0$에 대하여 $p^n$인 군을 $p$-군이라 부른다. 이제 $G$가 유한군이고 $H$는 $G$의 부분군이라고 하자. 만약 $H$가 $p$-군이라면, $H$를 $G$의 &lt;b&gt;$p$-부분군 ( $p$-Subgroup )&lt;/b&gt;이라고 부른다. 또한, 만약 $H$가 $G$의 극대 $p$-부분군인 경우, $H$를 $G$의 &lt;b&gt;쉴로브 $p$-부분군 ( Sylow $p$-Subgroup )&lt;/b&gt;이라고 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 본격적으로 쉴로브 정리를 다뤄보자. 쉴로브 정리는 위에서 언급했듯이 총 3개의 정리로 이루어졌으므로 본문을 총 세 개의 파트로 쪼개서 작성해보려 한다. 아래의 목차를 클릭하면 해당 부분을 볼 수 있다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now, let's deal with the Sylow theorems. As mentioned above, the Sylow theorems consist of three theorems, so we will divide the text into three parts. Click on the table of contents below to see those parts.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;h3 data-ke-size=&quot;size23&quot;&gt;&lt;a name=&quot;toc&quot;&gt;&lt;/a&gt;&lt;b&gt;Table of Contents&lt;/b&gt;&lt;/h3&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;ul style=&quot;list-style-type: disc;&quot; data-ke-list-type=&quot;disc&quot;&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Part1&quot;&gt;Part 1. The 1st Theorem&lt;/a&gt;&lt;/li&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Part2&quot;&gt;Part 2. The 2nd Theorem&lt;/a&gt;&lt;/li&gt;
&lt;li data-ke-style=&quot;style3&quot;&gt;&lt;a style=&quot;font-size: 110%;&quot; href=&quot;#Part3&quot;&gt;Part 3. The 3rd Theorem&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;/ul&gt;
&lt;/ul&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Part1&quot;&gt;&lt;/a&gt;&lt;b&gt;Part 1.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Before proving the 1st theorem of the Sylow theorems, let's show a lemma to prove it.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;쉴로브 정리의 첫 번째 정리를 증명하기에 앞서, 이를 보이기 위한 보조정리 하나를 보이자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Lemma1&quot;&gt;&lt;/a&gt;&lt;b&gt;Lemma 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a finite abelian group of order $m$, let $p$ be a prime number dividing $m$. Then $G$ has a subgroup of order $p$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;유한 아벨군 $G$의 위수가 $m$이고 소수 $p$가 $m$을 나눈다고 하자. 그러면 $G$는 위수가 $p$인 부분군을 가진다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We first prove by induction that if a finite abelian group $G$ has &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/193#ExponentOfGroup&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;exponent&lt;/b&gt;&lt;/a&gt; $n$ then the order of $G$ divides some power of $n$. Let $b \in G$, $b \neq 1$, and let $H$ be the &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/193&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;cyclic sugroup generated by $b$&lt;/b&gt;&lt;/a&gt;. (At this point, $1$ means an identity element of $G$.) Then the order of $H$ divides $n$ since $b^n = 1$, and $n$ is a multiple of exponent for $G/H$. Hence the order of $G/H$ divides a power of $n$ by induction hypothesis, and consequently so does the order of $G$ because &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/140&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;$\left( G : \mathbf{1} \right) = \left( G : H \right) \left( H : \mathbf{1} \right)$&lt;/a&gt;.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $G$ have order divisible by $p$. By what we have just seen, there exists $x \in G$ whose order is divisible by $p$. Let this order be $ps$ for some integer $s$. Then $x^s \neq 1$ and obviously $x^s$ has order $p$, and generates a subgroup of order $p$, as was to be shown.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;보조정리를 증명하기에 앞서 &quot;만약 $G$가 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/193#ExponentOfGroup&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;지수&lt;/b&gt;&lt;/a&gt; $n$을 가지면 $G$의 위수는 $n$의 한 거듭제곱을 나눈다.&quot;는 명제를 귀납법을 이용하여 보이자. 항등원이 아닌 $b \in G$와 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/193&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;$b$로부터 생성된 순환부분군&lt;/b&gt;&lt;/a&gt; $H$를 생각하자. 그러면 $b^n$이 항등원이라는 사실로부터 $H$의 위수가 $n$을 나눈다는 것을 알 수 있으며, $n$이 $G/H$의 지수의 배수가 됨을 알 수 있다. 귀납가정에 의해 $G/H$의 위수가 $n$의 한 거듭제곱을 나누므로 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/140&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;$\left( G : \mathbf{1} \right) = \left( G : H \right) \left( H : \mathbf{1} \right)$&lt;/a&gt;에 의해 $G$의 위수 역시 $n$의 한 거듭제곱을 나눈다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $G$의 위수가 $p$로 나누어진다고 하자. 그러면 방금 보인 사실로부터, $p$가 $\operatorname{ord} x$를 나누는 $x \in G$가 존재한다는 사실을 알 수 있으며, $x$의 위수를 $ps$라고 하면 $x^s$가 항등원이 아니며, $x^s$의 위수가 $p$가 된다는 사실을 쉽게 알 수 있다. 따라서 $x^s$로부터 생성된 순환부분군 $\left&amp;lt; x^s \right&amp;gt;$의 위수가 $p$이며, 따라서 보조정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 1. The 1st Theorem of Sylow Theorems&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a finite group and $p$ be a prime factor with multiplicity $n$ of the order of $G$. Then there exists a Sylow $p$-subgroup of $G$, of order $p^n$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;유한군 $G$와 $G$의 위수의 소인수 $p$를 생각하자. 만약 $p^n$이 $G$의 위수를 나누고 $p^{n+1}$은 $G$의 위수를 나누지 않는다면, 위수가 $p^n$인 $G$의 쉴로브 $p$-부분군이 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We will prove by induction on the order of $G$. If the order of $G$ is a prime, our assertion is obvious. We now assume given a finite group $G$, and assume that the theorem proved for all groups of order smaller than that of $G$. If there exists a proper subgroup $H$ of $G$ whose index is relatively prime to $p$, then a Sylow $p$-subgroup of $H$ will also be one of $G$, and our assertion follows by induction. We may therefore assume that every proper subgroup of $G$ has an index divisible by $p$. We now let $G$ act on itself by conjugation. From the fact that the set of the &lt;a href=&quot;https://chocobear.tistory.com/202#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;orbits under $G$&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; is a partition of $G$, we obtain $\left( G : \mathbf{1} \right) = \left( Z : \mathbf{1} \right) + \sum \left( G : G_x \right)$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Here, $Z$ is then center of $G$, and the term $\left( Z : \mathbf{1} \right)$ corresponds to the orbits having one element, namely the element of $Z$. The sum of the right is taken over the other orbits, and each index $\left( G : G_x \right)$ is then $&amp;gt;1$, hence $\sum \left( G : G_x \right)$ is divisible by $p$. Since $\left( G : \mathbf{1} \right)$ is divisible by $p$, $\left( Z : \mathbf{1} \right)$ is also divisible by $p$, hence $G$ has a non-trivial center.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $Z$ is a finite abelian group, by the &lt;a href=&quot;#Lemma1&quot;&gt;&lt;b&gt;Lemma 1&lt;/b&gt;&lt;/a&gt;, $Z$ has an element $a \in Z$ whose order is $p$. Let $H$ be the cyclic subgroup generated by $a$. Since $H$ is a subgroup of the center of $G$, $H$ is normal in $G$. Let $f : G \to G/H$ be a canonical map and let $p^n$ be the highest power of $p$ dividing the order of $G$. Then $p^{n-1}$ divides $\left( G : H \right)$. Let $K'$ be a Sylow $p$-subgroup of $G/H$. And let $K = f^{-1}(K')$. Then $H \subseteq K$ and $f$ maps $K$ onto $K'$. Hence we have an isomorphism $K/H \cong K'$ by the &lt;a href=&quot;https://chocobear.tistory.com/187#Part1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;first isomorphism theorem for groups&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;. Hence $K$ has order $p^{n-1}p = p^n$, as desired and this proves the theorem.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$의 위수에 대한 귀납법을 통해 정리를 증명할 것이다. 만약 $G$의 위수가 소수라면, 정리가 성립함은 자명하다. 이제 유한군 $G$가 주어졌다고 가정하고 위수가 $G$보다 작은 모든 유한군에 대해 정리가 성립한다고 가정하자. 만약 $G$의 진부분군 $H$의 지수가 $p$와 서로소라면 $H$의 위수가 $p^n$인 쉴로브 $p$-부분군은 $G$의 위수가 $p^n$인 쉴로브 $p$-부분군이 될 것이며, 그러한 부분군은 귀납가정에 의해 존재함을 알 수 있다. 따라서 이제 $G$의 모든 진부분군의 지수가 $p$의 배수라고 가정하고 $G$가 자기자신 위로 켤레로서 작용한다고 하자.&amp;nbsp; 그러면 &lt;a href=&quot;https://chocobear.tistory.com/202#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;$G$ 아래의 모든 궤도&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;의 집합이 $G$의 분할이 된다는 사실로부터 $\left( G : \mathbf{1} \right) = \left( Z : \mathbf{1} \right) + \sum \left( G : G_x \right)$를 얻을 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;여기서, $Z$는 $G$의 center이며, 항 $\left( Z : \mathbf{1} \right)$는 단원소 궤도들에 대응된다. 또한 우변의 시그마에서 각 지수 $\left( G : G_x \right)$는 원소가 하나 이상인 나머지 궤도들에 대응되며, 따라서 $\sum \left( G : G_x \right)$는 $p$로 나누어 떨어진다. $\left( G : \mathbf{1} \right)$가 $p$로 나누어 떨어지므로 $\left( Z : \mathbf{1} \right)$ 역시 $p$로 나누어 떨어지며, 따라서 $G$는 비자명 center를 가진다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$Z$가 유한 아벨군이라는 사실로부터&amp;nbsp;&lt;b&gt;Lemma 1&lt;/b&gt;에 의해 $Z$는 위수가 $p$인 원소 $a \in Z$를 가진다. $H$를 $a$로부터 생성된 순환 부분군이라고 하자. 그러면 $H$가 $G$의 center의 부분군이므로 $H$는 $G$의 정규부분군임을 알 수 있으며, 따라서 자연스러운 대응관계 $f : G \to G/H$가 존재한다. 그러면 귀납가정에 의해 위수가 $p^{n-1}$인 $G/H$의 쉴로브 $p$-부분군 $K'$이 존재하며, $K$를 $K = f^{-1}(K')$와 같이 정의하면 &lt;a href=&quot;https://chocobear.tistory.com/187#Part1&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;군에서의 제1동형 정리&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 의해 $K/H \cong K'$임을 알 수 있다. 따라서 $K$의 위수는 $p^{n-1}p = p^n$이며, 이로 인해 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Part2&quot;&gt;&lt;/a&gt;&lt;b&gt;Part 2.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Before proving the 2nd theorem of the Sylow theorems, let's show a lemma to prove it.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;쉴로브 정리의 두 번째 정리를 증명하기에 앞서, 이를 증명하기 위한 보조정리 하나를 보이자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Lemma2&quot;&gt;&lt;/a&gt;&lt;b&gt;Lemma 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $H$ be a finite $p$-group, let $\Omega$ be a finite set acted on by $H$, and let $\Omega_0$ denote the set of points of $\Omega$ that are fixed under the action of $H$. Then $\lvert \Omega \rvert \equiv \lvert \Omega_0 \rvert \left( \operatorname{mod} p \right)$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$H$가 $p$-군이고 $\Omega$가 유한 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;$H$-Set&lt;/b&gt;&lt;/a&gt;이며 $\Omega_0$은 $\Omega$ 위로의 $H$의 작용의 고정점의 집합이라고 하자. 그러면 $\lvert \Omega \rvert \equiv \lvert \Omega_0 \rvert \left( \operatorname{mod} p \right)$이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Any element $x \in \Omega$ not fixed by $H$ will lie in an orbit of order $\lvert H \rvert / \lvert H_x \rvert$ (where $H_x$ denotes the &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/202&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;stabilizer&lt;/b&gt;&lt;/a&gt;), which is a multiple of $p$ by assumption. The result follows immediately by writing $\lvert \Omega \rvert$ as the sum of $\lvert Hx \rvert$ over all distinct orbits $Hx$ and reducing mod $p$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;고정점이 아닌 $x \in \Omega$를 생각하자. 그러면 $x$는 위수가 $\lvert H \rvert / \lvert H_x \rvert$인 궤도의 원소임이 자명하다. 단, 여기서 $H_x$는 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/202&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;안정자군&lt;/b&gt;&lt;/a&gt;을 나타내며, 따라서 $x$는 위수가 $p$의 배수인 궤도의 원소가 된다. 따라서 $\lvert \Omega \rvert$를 모든 궤도 $Hx$에 대하여 $\lvert Hx \rvert$의 합으로 바꾸어 쓰면 보조정리가 증명됨을 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem2&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 2. The 2nd Theorem of Sylow Theorems&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $H$ is a $p$-subgroup of $G$ and $P$ is a Sylow $p$-subgroup of $G$, then there exists an element $g$ in $G$ such that $g^{-1}Hg \leq P$. In particular, all Sylow $p$-subgroups of $G$ are conjugate to each other, that is, if $H$ and $K$ are Sylow $p$-subgroups of $G$, then there exists an element $g$ in $G$ with $g^{-1}Hg = K$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$H$가 $G$의 $p$-부분군이고 $P$가 $G$의 쉴로브 $p$-부분군이라고 하자. 그러면 $g^{-1}Hg \leq P$이도록 하는 $g \in G$가 존재한다. 더 나아가서, $G$의 모든 쉴로브 $p$-부분군은 켤레관계에 있다. 즉, $H$와 $K$가 $G$의 쉴로브 $p$-부분군이면 $g^{-1}Hg = K$이도록 하는 $g \in G$가 존재한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\Omega$ be the set of left cosets of $P$ in $G$ and let $H$ act on $\Omega$ by left multiplication. Applying the &lt;a href=&quot;#Lemma2&quot;&gt;&lt;b&gt;Lemma 2&lt;/b&gt;&lt;/a&gt; to $H$ on $\Omega$, we see $\lvert \Omega_0 \rvert \equiv \lvert \Omega \rvert \equiv \left( G : P \right) \left( \operatorname{mod} p \right)$. Now $p \nmid \left( G : P \right)$ by definition so $p \nmid \lvert \Omega_0 \rvert$, hence in particular $\lvert \Omega_0 \rvert \neq 0$ so there exists some $gP \in \Omega_0$. With this $gP$, we have $hgP = gP$ for all $h \in H$, so $HgP = gP$ and therefore $g^{-1}Hg \leq P$. Furthermore, if $H$ is a Sylow $p$-subgroup, then $\lvert g^{-1}Hg \rvert = \lvert H \rvert = \lvert P \rvert$ so that $g^{-1}Hg = P$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$에서 $P$의 좌잉여류의 집합을 $\Omega$라고 하고 $H$가 $\Omega$에 왼쪽 곱셈으로서 작용한다고 하자. $\Omega$와 $H$에 대하여 &lt;a href=&quot;#Lemma2&quot;&gt;&lt;b&gt;Lemma 2&lt;/b&gt;&lt;/a&gt;를 적용하면 $\lvert \Omega_0 \rvert \equiv \lvert \Omega \rvert \equiv \left( G : P\right) \left( \operatorname{mod} p \right)$임을 알 수 있다. 그런데 $P$의 정의에 의해 $p \nmid \left( G : P \right)$임이 자명하므로 $\lvert \Omega_0 \rvert \neq 0$임을 알 수 있다. 따라서 어떤 $ g \in G$에 대하여 $gP \in \Omega_0$이 성립하며, 따라서 [임의의 $h \in H$에 대해 $hgP = gP$임을 알 수 있다. 이로부터 $HgP = gP$임을 알 수 있으며, 따라서 $g^{-1}Hg \leq P$를 얻는다. 더 나아가서, $H$가 쉴로브 $p$-부분군이라면, $\lvert g^{-1}Hg \rvert = \lvert H \rvert = \lvert P \rvert$이며 이로 인해 $g^{-1}Hg = P$임을 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Part3&quot;&gt;&lt;/a&gt;&lt;b&gt;Part 3. The 3rd Theorem of Sylow Theorems&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;As the above two theorems give information about the order of the Sylow p-subgroup, the last theorem gives information about the number of the Sylow p-subgroup.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위의 두 정리가 쉴로브 p-부분군의 위수에 대한 정보를 준다면, 이번 정리는 쉴로브 p-부분군의 개수에 대한 정보를 준다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem3&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $q$ denote the order of any Sylow $p$-subgroup $P$ of a finite group $G$. Let $n_p$ denote the number of Sylow $p$-subgroups of $G$. Then:&lt;br /&gt;(a) $n_p = \left( G : N_G(P) \right)$,&lt;br /&gt;(b) $n_p$ divides $\lvert G \rvert / q$, and&lt;br /&gt;(c) $n_p \equiv 1 \left( \operatorname{mod} p \right)$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;유한군 $G$의 쉴로브 $p$-부분군 $P$에 대하여 $P$의 위수가 $q$라고 하고 $n_p$가 $G$의 쉴로브 $p$-부분군의 개수를 나타낸다고 하자. 그러면 다음이 성립한다.&lt;br /&gt;(a) $n_p = \left( G : N_G(P) \right)$&lt;br /&gt;(b) $n_p$가 $\lvert G \rvert / q$를 나눈다.&lt;br /&gt;(c) $n_p \equiv 1 \left( \operatorname{mod} p \right)$&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\Omega$ be the set of all Sylow $p$-subgroups of $G$ and let $G$ act on $\Omega$ by conjugation. Let $P \in \Omega$ be a Sylow $p$-subgroup. By &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;, the orbit of $P$ has size $n_p$, so by the &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/202#Theorem3&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;orbit-stabilizer theorem&lt;/b&gt;&lt;/a&gt; $n_p = \left( G : G_P \right)$. For this group action, the stabilizer $G_P$ is given by $\{ g \in G \;|\; gPg^{-1} = P \} = N_G(P)$, the normalizer of $P$ in $G$. Thus, $n_p = \left( G : N_G(P) \right)$, and it follows that this number is a divisor of $\left( G : P \right) = \lvert G \rvert / q$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let $P$ act on $\Omega$ by conjugation, and again let $\Omega_0$ denote the set of fixed points of this action. Let $Q \in \Omega_0$ and observe that then $Q = xQx^{-1}$ for all $x \in P$ so that $P \leq N_G(Q)$. By &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;, $P$ and $Q$ are conjugate in $N_G(Q)$ in particular, and $Q$ is normal in $N_G(Q)$, so then $P=Q$. It follows that $\Omega_0 = \{ P \}$ so that by the &lt;a href=&quot;#Lemma2&quot;&gt;&lt;b&gt;Lemma 2&lt;/b&gt;&lt;/a&gt;, $\lvert \Omega \rvert \equiv \lvert \Omega_0 \rvert = 1 \left( \operatorname{mod} p \right)$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$의 모든 쉴로브 $p$-부분군의 집합을 $\Omega$라고 하고 $G$가 $\Omega$ 위로 켤레로서 작용한다고 하자. 그리고 $P \in \Omega$가 $G$의 쉴로브 $p$-부분군이라고 하자. &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;에 의해 $P$의 궤도는 $n_p$개의 원소를 가지므로 &lt;a href=&quot;https://chocobear.tistory.com/202#Theorem3&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;궤도-안정자군 정리&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 의해 $n_p = \left( G : G_P \right)$가 성립한다. 군의 작용에 대하여 안정자군 $G_P$는 $\{ g \in G \;|\; gPg^{-1} = P \} = N_G(P)$로 주어지므로 $P$의 normalizer임을 알 수 있다. 따라서 $n_p = \left( G : N_G(P) \right)$이며, 이는 곧 $n_p$가 $\left( G : P \right) = \lvert G \rvert / q$의 약수임을 함의한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $P$가 $\Omega$ 위로 켤레로서 작용한다고 하고 $\Omega_0$이 이 작용의 고정점의 집합이라고 하자. 이때 $Q \in \Omega_0$이라고 하면 임의의 $x \in P$에 대하여 $Q = xQx^{-1}$이 성립하므로 $P \leq N_G(Q)$임을 알 수 있다. &lt;a href=&quot;#Theorem2&quot;&gt;&lt;b&gt;Theorem 2&lt;/b&gt;&lt;/a&gt;에 의해 $P$와 $Q$는 $N_G(Q)$에서 켤레관계이며, normalizer의 정의에 의해 $Q$가 $N_G(Q)$의 정규부분군이므로 $P = Q$임을 알 수 있다. 이는 곧 $\Omega_0 = \{ P \}$를 의미하며, 따라서&amp;nbsp;&lt;a href=&quot;#Lemma2&quot;&gt;&lt;b&gt;Lemma 2&lt;/b&gt;&lt;/a&gt;에 의해 $\lvert \Omega \rvert \equiv \lvert \Omega_0 \rvert = 1 \left( \operatorname{mod} p \right)$임을 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Sylow Subgroups</category>
      <category>Sylow Theorems</category>
      <category>쉴로브 부분군</category>
      <category>쉴로브 정리</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/204</guid>
      <comments>https://chocobear.tistory.com/204#entry204comment</comments>
      <pubDate>Sat, 19 Feb 2022 18:01:56 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물세 번째 이야기  |  대칭군과 교대군 ( Symmetric Group &amp;amp; Alternating Group )</title>
      <link>https://chocobear.tistory.com/203</link>
      <description>&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 대칭군과 교대군에 대한 이야기를 해보려 한다. 대칭군과 교대군은 특수한 구조를 지니는 군을 가르키는 말이며, 몇 가지 중요한 성질들을 가지고 있기 때문에 꽤 중요하게 다뤄지는 대상이다. 먼저, 대칭군이 무엇인지 알아보도록 하자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, I would like to talk about symmetric groups and alternating groups. Symmetric groups and alternating groups refer to groups with special structures, and are considered quite important because they have several important properties. First, let's find out what a symmetric group is.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. 대칭군 ( Symmetric Group )&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. The Symmetric Group&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;The &lt;b&gt;symmetric group&lt;/b&gt; $S_n$ is the group of permutations of a set with $n$ elements, whose group operation is the composition of functions. At this point, permutations of a set $S$ refer to bijections from the set $S$ to itself. The set with $n$ elements may be taken to the set of integers $J_n = \{ 1,2,\cdots,n \}$.&lt;br /&gt;Let $S_n$ be the symmetric group on the set $J_n$. Given any $\sigma \in S_n$, and any integer $i$, $1 \leq i \leq n$, we may form the orbit of $i$ under the cyclic group generated by $\sigma$. Such an &lt;a href=&quot;https://chocobear.tistory.com/202#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;orbit&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; is called a &lt;b&gt;cycle&lt;/b&gt; for $\sigma$, and may be written $\left( i_1 ~ i_2 ~ \cdots ~ i_r \right)$, so $\sigma(i_1) = i_2, \cdots, \sigma(i_{r-1}) = i_r, \sigma(i_r) = i_1$. Then $J_n$ may be decomposed into a disjoint union of orbits for the cyclic group generated by $\sigma$, and therefore into disjoint cycles. Thus the effect of $\sigma$ on $J_n$ is represented by a product of disjoint cycles.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;&lt;b&gt;대칭군 ( Symmetric Group )&lt;/b&gt; $S_n$은 $n$개의 원소로 이루어진 집합의 치환의 군으로 정의된다. 이때, 군에 부여되는 이항연산은 함수의 합성이며, $n$개의 원소로 이루어진 집합은 일반적으로 $J_n = \{ 1,2, \cdots, n \}$을 사용한다. 여기서, 집합 $S$의 치환 ( permutation )이라 함은, $S$에서 $S$ 자신으로 가는 일대일 대응을 말한다. 순열과 같은 의미이며, 특히 대수학 분야에서 이 대상을 치환이라고 부른다.&lt;br /&gt;$S_n$을 집합 $J_n$ 위에서 정의된 대칭군이라고 하자. 그러면 임의의 $\sigma \in S_n$과 $1 \leq i \leq n$인 임의의 정수 $i$에 대하여 $\sigma$로부터 생성된 순환군 아래에서 $i$의 &lt;a href=&quot;https://chocobear.tistory.com/202#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;궤도&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;를 형성할 수 있을 것이다. 이렇게 형성된 궤도를 $\sigma$의 &lt;b&gt;Cycle&lt;/b&gt;이라고 하며, 이를 $\left( i_1 ~ i_2 ~ \cdots ~ i_r \right)$과 같이 나타낸다. 당연하지만, $\left( i_1 ~ i_2 ~ \cdots ~ i_r \right)$가 $\sigma$의 cycle인 경우, $\sigma(i_1) = i_2 , \cdots , \sigma(i_{r-1}) = i_r , \sigma(i_r) = i_1$이다. 이와 같이 cycle이 정의되면, $J_n$은 $\sigma$의 cycle의 disjoint union이 됨은 자명하다. 따라서 $\sigma$는 서로소인 cycle들의 곱으로 나타낼 수 있다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;If a symmetric group is defined as described above, a 'sign' may be assigned to each element of the symmetric group, that is, each permutation. Let's look at the theorem and definition below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위와 같이 대칭군을 정의하면, 대칭군의 각 원소 즉, 각 순열에 '부호'를 부여할 수 있다. 아래의 정리와 정의를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;There is a unique homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$ such that for every transposition $\tau$ we have $\varepsilon ( \tau ) = -1$. At this point, a transposition is a permutation such that exchanges two elements and keeps all other elements.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;임의의 호환 $\tau$에 대하여 $\varepsilon ( \tau ) = -1$이 성립하도록 하는 group-homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$이 유일하게 존재한다. 이때, 호환 ( transposition )이라 함은, 정확히 두 원소를 교환하고 나머지 원소는 그대로 두는 치환을 말한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;If there is a homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$ such that for every transposition $\tau$ we have $\varepsilon(\tau) = -1$, then the homomorphism $\varepsilon$ is unique since every permutation can be represented by a product of transpositions. So we have to show that there is a homomorphism satisfying the above conditions.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\Delta$ be the function $\Delta \left( x_1 , \cdots , x_n \right) = \displaystyle \prod_{i &amp;lt; j} \left( x_j - x_i \right)$, the product being taken for all pairs of integers $i,j$ satisfying $1 \leq i &amp;lt; j \leq n$. Let $\tau$ be a transposition, interchanging the two integers $r$ and $s$. Say $r &amp;lt; s$. We wish to determine $\tau \Delta \left( x_1 , \cdots , x_n \right) = \displaystyle \prod_{i &amp;lt; j} \left( x_{\tau(i)} - x_{\tau(j)} \right)$ as a &lt;a href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;group action&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;. For one factor involving $i=r$, $j=s$, we see that $\tau$ changes the factor $\left( x_s - x_r \right)$ to $-\left( x_s - x_r \right)$. All other factors can be considered in pairs as follows: $$ \begin{array}{l} ( x_k - x_s ) ( x_k - x_r ) \text{ if } k &amp;gt; s, \\ ( x_s - x_k ) ( x_k - x_r ) \text{ if } r &amp;lt; k &amp;lt; s, \\ ( x_s - x_k ) ( x_r - x_k ) \text{ if } k &amp;lt; r \end{array} $$&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Each one of these pairs remains unchanged when we apply $\tau$. Hence we see that $\tau \Delta = - \Delta$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\varepsilon ( \sigma )$ be $1$ or $-1$ such that $\sigma \Delta = \varepsilon(\sigma) \Delta$ for a permutation $\sigma$. $\sigma \Delta$ is defined as the case of transpositions, of course. Since $(\sigma \tau)\Delta = \sigma(\tau\Delta)$, it follos at once that $\varepsilon$ is a homomorphism, and the theorem is proved.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;만약 임의의 호환 $\tau$에 대하여 $\varepsilon(\tau) = -1$이 성립하도록 하는 homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$이 존재한다면, 모든 치환은 유한한 수의 호환의 곱으로 나타낼 수 있으므로 이 조건을 만족하는 homomorphism은 유일하다. 따라서 이런 조건을 만족하는 homomorphism이 존재한다는 것을 보이면 충분하다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 함수 $\Delta$를 $\Delta \left( x_1 , \cdots , x_n \right) = \displaystyle \prod_{i &amp;lt; j } \left( x_j - x_i \right)$로 정의하자. 이때, $i,j$는 $1 \leq i &amp;lt; j \leq n$을 만족하는 두 정수이다. 이제 $\tau$가 $r$와 $s$를 교환하는 호환이라고 하고 $r &amp;lt; s$라고 하자. 이때 $\tau \Delta \left( x_1 , \cdots , x_n \right) = \displaystyle \prod_{i &amp;lt; j} \left( x_{\tau(i)} - x_{\tau(j)} \right)$를 일종의 &lt;a href=&quot;https://chocobear.tistory.com/201&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Group Action&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;으로 바라보자. 그러면 $\tau$가 $\left( x_s - x_r \right)$를 $-\left( x_s - x_r \right)$로 바꾼다는 것을 알 수 있다. 또한, 그 외의 모든 요소들은 다음의 세 가지 경우로 분류할 수 있다. $$ \begin{array}{l} ( x_k - x_s ) ( x_k - x_r ) \text{ if } k &amp;gt; s, \\ ( x_s - x_k ) ( x_k - x_r ) \text{ if } r &amp;lt; k &amp;lt; s, \\ ( x_s - x_k ) ( x_r - x_k ) \text{ if } k &amp;lt; r \end{array} $$&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이때, 위의 세 가지는 모두 $\tau$가 적용되어도 변하지 않는다는 사실을 쉽게 알 수 있으며, 따라서 $\tau \Delta = - \Delta$임을 알 수 있다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 치환 $\sigma$에 대하여 $\varepsilon ( \sigma )$에 $\sigma \Delta = \varepsilon(\sigma) \Delta$를 만족하도록 $1$ 또는 $-1$의 값을 부여하자. 당연하지만, 위에서의 $\sigma \Delta$는 호환에 대한 정의와 같은 방식으로 정의된다. 그러면 $(\sigma \tau)\Delta = \sigma(\tau\Delta)$라는 사실로부터 $\varepsilon$이 homomorphism이 된다는 사실을 즉시 알 수 있으며, 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition2&quot;&gt;&lt;/a&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 2. 치환의 부호 ( Sign of Permutation )&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 2. The Sign of Permutation&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;ko&quot;&gt;&lt;a href=&quot;#Theorem1&quot;&gt;&lt;b&gt;Theorem 1&lt;/b&gt;&lt;/a&gt;에서 정의된 homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$과 치환 $\sigma \in S_n$에 대하여 $\varepsilon(\sigma)$를 $\sigma$의 &lt;b&gt;부호 ( Sign )&lt;/b&gt;라고 부르며, 부호가 $1$인 치환을 &lt;b&gt;짝치환 ( Even Permutation )&lt;/b&gt;, 부호가 $-1$인 치환을 &lt;b&gt;홀치환 ( Odd Permutation )&lt;/b&gt;이라고 부른다. 또한, $S_n$의 원소 중 짝치환만을 모두 모아놓은 $S_n$의 부분군을 &lt;b&gt;교대군 ( Alternating Group )&lt;/b&gt;이라고 부르며, $A_n$으로 나타낸다.&lt;/div&gt;
&lt;div class=&quot;en&quot;&gt;For the homomorphism $\varepsilon : S_n \to \{ 1 , -1 \}$ defined in &lt;a href=&quot;#Theorem1&quot;&gt;&lt;b&gt;Theorem 1&lt;/b&gt;&lt;/a&gt; and a permutation $\sigma \in S_n$, $\varepsilon(\sigma)$ is called the &lt;b&gt;sign&lt;/b&gt; of $\sigma$. As a manner of terminology, we call $\sigma$ &lt;b&gt;even&lt;/b&gt; if $\varepsilon(\sigma)$ is $1$, and &lt;b&gt;odd&lt;/b&gt; if $\varepsilon(\sigma) = -1$. The even permutations constitute the kernel of $\varepsilon$, which is called the &lt;b&gt;alternating group&lt;/b&gt; $A_n$.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we know what symmetric group and alternating group are, let's find out the properties of them.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 대칭군과 교대군이 무엇인지 알았으니 이들의 성질에 대해 알아보도록 하자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $n \geq 5$, then $S_n$ is not &lt;a href=&quot;https://chocobear.tistory.com/194#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;solvable&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$n \geq 5$라면 $S_n$은 &lt;a href=&quot;https://chocobear.tistory.com/194#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Solvable&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;하지 않다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We shall first prove that if $H,N$ are two subgroups of $S_n$ where $n \geq 5$ such that $N \subseteq H$ and $N$ is normal in $H$, if $H$ contains every $3$-cycle and $H/N$ is abelian, then $N$ contains every $3$-cycle. To see this, let $i,j,k,r,s$ be five distinct integers in $J_n$, and let $\sigma = \left( i ~ j ~ k \right)$ and $\tau = \left( k ~ r ~ s \right)$. Then a direct computation gives their commutator $\sigma\tau\sigma^{-1}\tau^{-1} = \left( r ~ k ~ i \right)$. Since the choice of $i,j,k,r,s$ was arbitrary, we see that the cycles $\left( r ~ k ~ i \right)$ all lie in $N$ for all choices of distinct $r,k,i$, thereby proving what we wanted.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now suppose that we have a &lt;a href=&quot;https://chocobear.tistory.com/194&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;tower of subgroups&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $S_n = H_0 \supseteq H_1 \supseteq \cdots \supseteq H_m = \mathbf{1}$ which is abelian. Since $S_n$ has every $3$-cycle, we conclude that $H_1$ has every $3$-cycle. By induction, we conclude that $H_m = \mathbf{1}$ has every $3$-cycle, which is impossible, thus proving the theorem.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;정리의 증명에 앞서, 만약 $H,N$이 $S_n$의 두 부분군이고, $n$이 $5$ 이상이며, $N$이 $H$의 정규부분군이며, $H$가 모든 $3$-cycle을 원소로 가지고 $H/N$이 아벨군이라면, $N$ 역시 모든 $3$-cycle을 원소로 가진다는 사실을 보일 것이다. 이를 보이기 위해, $i,j,k,r,s$를 $J_n$의 서로 다른 $5$개의 정수라고 하고 $\sigma = \left( i ~ j ~ k \right)$, $\tau = \left( k ~ r ~ s \right)$라고 하자. 이제 직접 계산을 해보면 $\sigma$와 $\tau$의 교환자가 $\sigma\tau\sigma^{-1}\tau^{-1} = \left( r ~ k ~ i \right)$임을 알 수 있다. $i,j,k,r,s$의 선택이 임의적이라는 사실로부터, 우리는 임의의 서로 다른 세 정수 $r,k,i$에 대하여 cycle $\left( r ~ k ~ i \right)$가 $N$의 원소가 됨을 알 수 있으며, 따라서 목표했던 명제가 증명된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 &lt;a href=&quot;https://chocobear.tistory.com/194&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Normal Tower&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $S_n = H_0 \supseteq H_1 \supseteq \cdots \supseteq H_m = \mathbf{1}$가 abelian tower라고 가정하자. 그러면 $S_n$이 모든 $3$-cycle을 원소로 가진다는 사실로부터 $H_1$ 역시 모든 $3$-cycle을 원소로 가진다는 사실을 알 수 있다. 같은 논리를 유한 번 반복 적용하면, $H_m = \mathbf{1}$ 또한 모든 $3$-cycle을 원소로 가진다는 사실을 알 수 있는데, 이는 자명하게 모순이다. 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem3&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $n \geq 3$, $A_n$ is generated by the $3$-cycles.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;만약 $n \geq 3$이라면, $A_n$은 $3$-cycle들로부터 생성된다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Consider the product of two transposition $\left( i ~ j \right) \left( r ~ s \right)$. If they have an element in common, then the product is either the identity permutaion or a $3$-cycle. If they don't have an element in common, then $\left( i ~ j \right) \left( r ~ s \right) = \left( i ~ j ~ r \right) \left( j ~ r ~ s \right)$, so the product of two transposition is also a product of $3$-cycles. Since an even permutation is a product of even transpositions, we are done.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;두 호환의 곱 $\left( i ~ j \right) \left( r ~ s \right)$를 생각하자. 만약 이 두 호환이 같은 요소를 공유한다면, 그 둘의 곱은 항등 변환이거나 $3$-cycle이 된다. 만약 이 두 호환이 같은 요소를 공유하지 않는다면, $\left( i ~ j \right) \left( r ~ s \right) = \left( i ~ j ~ r \right) \left( j ~ r ~ s \right)$이므로 두 호환의 곱은 두 $3$-cycle의 곱임을 알 수 있다. 모든 짝치환은 짝수 개의 호환의 곱으로 표현 가능하므로, 정리의 증명이 완료된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem4&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 4.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $n \geq 5$, all $3$-cycles are conjugate in $A_n$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;만약 $n \geq 5$라면, 임의의 두 $3$-cycle은 $A_n$에서 켤레관계이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $\gamma$ be a permutation in $S_n$ and $\left( i_1 ~ i_2 ~ \cdots ~ i_m \right)$ be a cycle. Then $\gamma \left( i_1 ~ i_2 ~ \cdots ~ i_m \right) \gamma^{-1} = \left( \gamma(i_1) ~ \gamma(i_2) ~ \cdots ~ \gamma(i_m) \right)$. Take two $3$-cycles $\left( i ~ j ~ k \right)$ and $\left( i' ~ j' ~ k' \right)$. Then we can find a permutation $\gamma$ such that $\gamma(i) = i'$, $\gamma(j) = j'$, and $\gamma(k) = k'$. If $\gamma$ is an even permutation, then $\left( i ~ j ~ k \right)$ and $\left( i' ~ j' ~ k' \right)$ are conjugate in $A_n$ since $\gamma \left( i ~ j ~ k \right) \gamma^{-1} = \left( i' ~ j' ~ k' \right)$. If $\gamma$ is an odd permutation, then we can choose distinct integers $r,s$ in $J_n$ such that $\left( i ~ j ~ k \right)$ and $\left( r ~ s \right)$ have no common element. Since $\gamma \left( r ~ s \right) (i) = i'$, $\gamma \left( r ~ s \right) (j) = j'$, and $\gamma \left( r ~ s \right) (k) = k'$, $\gamma \left( r ~ s \right) \left( i ~ j ~ k \right) \left( r ~ s \right)^{-1} \gamma^{-1} = \left( i' ~ j' ~ k' \right)$. Therefore, all $3$-cycles are conjugate in $A_n$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$\gamma$가 $S_n$의 원소이고 $\left( i_1 ~ i_2 ~ \cdots ~ i_m \right)$가 cycle이라고 하자. 그러면 $\gamma \left( i_1 ~ i_2 ~ \cdots ~ i_m \right) \gamma^{-1} = \left( \gamma(i_1) ~ \gamma(i_2) ~ \cdots ~ \gamma(i_m) \right)$임을 알 수 있다. 이제 두 $3$-cycle $\left( i ~ j ~ k \right)$와 $\left( i' ~ j' ~ k' \right)$를 잡자. 그러면 $\gamma(i) = i'$, $\gamma(j) = j'$, $\gamma(k) = k'$을 만족하는 치환 $\gamma \in S_n$를 잡을 수 있다. 만약 그런 $\gamma$가 짝치환이라면, $\gamma \left( i ~ j ~ k \right) \gamma^{-1} = \left( i' ~ j' ~ k' \right)$이므로 $\left( i ~ j ~ k \right)$와 $\left( i' ~ j' ~ k' \right)$는 $A_n$에서 켤레관계이다. 만약 그러한 $\gamma$가 홀치환이라면, $n \geq 5$라는 사실로부터 $i,j,k$와 구분되는 서로 다른 두 정수 $r,s \in J_n$를 잡을 수 있다. 이때, $\gamma \left( r ~ s \right) (i) = i'$, $\gamma \left( r ~ s \right) (j) = j'$, $\gamma \left( r ~ s \right) (k) = k'$이므로 $\gamma \left( r ~ s \right) \left( i ~ j ~ k \right) \left( r ~ s \right)^{-1} \gamma^{-1} = \left( i' ~ j' ~ k \right)$임을 알 수 있으며, 따라서 이 경우에도 $\left( i ~ j ~ k \right)$와 $\left( i' ~ j' ~ k' \right)$는 $A_n$에서 켤레관계이다. 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Corollary4_1&quot;&gt;&lt;/a&gt;&lt;b&gt;Corollary 4.1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If a normal subgroup $N$ of $A_n$ contains a $3$-cycle, then $N = A_n$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;만약 $N$이 $A_n$의 정규부분군이고 하나의 $3$-cycle을 원소로 가진다면 $N = A_n$이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;b&gt;&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since&amp;nbsp;&lt;a href=&quot;#Theorem4&quot;&gt;&lt;b&gt;Theorem 4&lt;/b&gt;&lt;/a&gt;, every $3$-cycle is in $N$. And also, every even permutation is in $N$ by&amp;nbsp;&lt;a href=&quot;#Theorem3&quot;&gt;&lt;b&gt;Theorem 3&lt;/b&gt;&lt;/a&gt;. Therefore, $N = A_n$ since $N \trianglelefteq A_n$ and $N \supseteq A_n$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;a href=&quot;#Theorem4&quot;&gt;&lt;b&gt;Theorem 4&lt;/b&gt;&lt;/a&gt;에 의해서 모든 $3$-cycle은 $N$의 원소이다. 또한,&amp;nbsp;&lt;a href=&quot;#Theorem3&quot;&gt;&lt;b&gt;Theorem 3&lt;/b&gt;&lt;/a&gt;에 의해 모든 짝치환이 $N$의 원소이다. 따라서 $N \trianglelefteq A_n$이고 $N \supseteq A_n$이므로 $N = A_n$이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 5.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $n \geq 5$ then the alternating group $A_n$ is &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/197#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;simple&lt;/b&gt;&lt;/a&gt;.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$n \geq 5$이면 교대군 $A_n$은 &lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/197#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;b&gt;단순군&lt;/b&gt;&lt;/a&gt;이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $N \trianglelefteq A_n$ be a nontrivial normal subgroup. We will show that $N=A_n$. The proof now proceeds by cases. In each cases, the normality of $N$ will allow us to reduce the proof to&amp;nbsp;&lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt; or to one of the previous cases:&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$N \trianglelefteq A_n$이 $A_n$의 비자명 정규부분군이라고 하자. 이제 $N = A_n$임을 보일 것이며, 증명은 총 다섯 가지의 케이스로 쪼개서 진행될 것이다. 각 케이스는 &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corrollary 4.1&lt;/b&gt;&lt;/a&gt;으로 환원되거나 다른 케이스로 환원되며, 따라서 정리가 증명되게 된다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Case 1.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose that there is a $\pi \in N$ that, when written as dijoint cycles, has a cycle of length at least $4$, say $\pi = \left( a_1 ~ a_2 ~ a_3 ~ a_4 ~ \cdots \right) \cdots$. Upon conjugation by $\left( a_1 ~ a_2 ~ a_3 \right) \in A_n$, we obtain $\pi' = \left( a_1 ~ a_2 ~ a_3 \right) \pi \left( a_3 ~ a_2 ~ a_1 \right) = \left( a_2 ~ a_3 ~ a_1 ~ a_4 ~ \cdots \right) \cdots$. Hence, $\pi' \in N$, and hence $\pi'\pi^{-1} = \left( a_1 ~ a_2 ~ a_4 \right) \in N$ also. Notice that the rest of the cycles cancel. By &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;, $N = A_n$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;길이가 $4$ 이상인 cycle을 가지는 $\pi \in N$가 존재한다고 가정해보자. 그러면 $\pi = \left( a_1 ~ a_2 ~ a_3 ~ a_4 ~ \cdots \right) \cdots$와 같이 표현할 수 있다. 이제 $\pi$에 $\left( a_1 ~ a_2 ~ a_3 \right) \in A_n$에 대한 켤레 연산을 취해주면 $\pi' = \left( a_1 ~ a_2 ~ a_3 \right) \pi \left( a_3 ~ a_2 ~ a_1 \right) = \left( a_2 ~ a_3 ~ a_1 ~ a_4 ~ \cdots \right) \cdots$를 얻을 수 있으며, $N$이 $A_n$의 정규부분군이므로 $\pi' \in N$임을 알 수 있다. 따라서 $\pi'\pi^{-1} = \left( a_1 ~ a_2 ~ a_4 \right) \in N$이 성립함을 알 수 있으며, &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;에 의해 $N = A_n$임을 알 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Case 2.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose that there is a $\pi \in N$ whose disjoint cycle decomposition has at least two cycles of length $3$, say $\pi = \left( a ~ b ~ c \right) \left( d ~ e ~ f \right) \cdots$. Conjugation by $\left( c ~ d ~ e \right) \in A_n$ implies that $N$ also contains $\pi' = \left( c ~ d ~ e \right) \pi \left( e ~ d ~ c \right) = \left( a ~ b ~ d \right)) \left( e ~ c ~ f \right) \cdots$. Hence, $N$ also contains $\pi'\pi = \left( a ~ d ~ c ~ b ~ f \right) \cdots$. This reduces the proof to&amp;nbsp;&lt;b&gt;Case 1&lt;/b&gt;.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번엔 길이가 $3$ 이상인 서로 다른 두 개 이상의 cycle을 가지는 $\pi \in N$가 존재한다고 가정해보자. 그러면 $\pi = \left( a ~ b ~ c \right) \left( d ~ e ~ f \right) \cdots$와 같이 표현할 수 있다. 이제 $\pi$에 $\left( c ~ d ~ e \right) \in A_n$에 대한 켤레 연산을 취해주면 $\pi' = \left( c ~ d ~ e \right) \pi \left( e ~ d ~ c \right) = \left( a ~ b ~ d \right) \left( e ~ c ~ f \right) \cdots$ 역시 $N$의 원소임을 알 수 있다. 따라서 $\pi'\pi = \left( a ~ d ~ c ~ b ~ f \right) \cdots \in N$임을 알 수 있고, 따라서 &lt;b&gt;Case 1&lt;/b&gt;의 상황으로 환원된다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Case 3.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose that there is a $\pi \in N$ such that the disjoint cycle decomposition has one cycle of length $3$ and transposition of even number (can be zero). Then $\pi\pi$ is a cycle of length $3$. By &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;, $N = A_n$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번엔 disjoint cycle decomposition이 정확히 한 개의 길이 $3$짜리 cycle과 짝수 개의 호환으로 이루어진 $\pi \in N$이 존재하는 경우를 생각해보자. 그러면 $\pi\pi$가 $3$-cycle이 됨이 자명하므로 &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;에 의해 $N = A_n$임을 알 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Case 4.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose that there is a $\pi \in N$ of the form $\pi = \left( a ~ b \right) \left( c ~ d \right)$. Conjugating by $\left( a ~ e ~ b \right)$ with $e$ distinct from $a$, $b$, $c$, and $d$ yeilds $\pi' = \left( a ~ e ~ b \right) \pi \left( b ~ e ~ a \right) = \left( a ~ e \right) \left( c ~ d \right)$. Then $\pi\pi' = \left( a ~ e ~ b \right)$. By &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;, $N = A_n$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번엔 두 호환의 곱인 $\pi \in N$가 존재한다고 가정하자. 그러면 $\pi = \left( a ~ b \right) \left( c ~ d \right)$와 같이 표현할 수 있다. 그러면 $n \geq 5$이므로 $a,b,c,d$와 다른 정수 $e \in J_n$가 존재할 것이며, 따라서 $\left( a ~ e ~ b \right) \in A_n$에 대한 켤레연산을 취할 수 있다. 따라서 $\pi' = \left( a ~ e ~ b \right) \pi \left( b ~ e ~ a \right) = \left( a ~ e \right) \left( c ~ d \right)$가 $N$의 원소이며, 따라서 $\pi\pi' = \left( a ~ e ~ b \right)$ 역시 $N$의 원소임을 알 수 있다. 따라서 &lt;a href=&quot;#Corollary4_1&quot;&gt;&lt;b&gt;Corollary 4.1&lt;/b&gt;&lt;/a&gt;에 의해 $N = A_n$임을 알 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Case 5.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose that $N$ contains a permutation of the form $\pi = \left( a_1 ~ b_1 \right) \left( a_2 ~ b_2 \right) \left( a_3 ~ b_3 \right) \left( a_4 ~ b_4 \right) \cdots$. This time we conjugate by $\left( a_2 ~ b_1 \right) \left( a_3 ~ b_2 \right)$. $\pi' = \left( a_2 ~ b_1 \right) \left( a_3 ~ b_2 \right) \pi \left( a_3 ~ b_2 \right) \left( a_2 ~ b_1 \right) = \left( a_1 ~ a_2 \right) \left( b_1 ~ a_3 \right) \left( b_2 ~ b_3 \right) \left( a_4 ~ b_4 \right) \cdots$. Then $\pi\pi' = \left( a_1 ~ b_2 ~ a_3 \right) \left( a_2 ~ b_1 ~ b_3 \right)$. This reduces the proof to&amp;nbsp;&lt;b&gt;Case 2&lt;/b&gt;.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 위에서 고려하지 않은 경우인 $\pi \in N$가 $4$ 이상의 짝수 개의 서로소인 호환의 곱으로 이루어진 경우를 가정하자. 그러면 $\pi = \left( a_1 ~ b_1 \right) \left( a_2 ~ b_2 \right) \left( a_3 ~ b_3 \right) \left( a_4 ~ b_4 \right) \cdots$와 같이 표현할 수 있다. 여기에 $\left( a_2 ~ b_1 \right) \left( a_3 ~ b_2 \right) \in A_n$에 대한 켤레연산을 취하면 $\pi' = \left( a_2 ~ b_1 \right) \left( a_3 ~ b_2 \right) \pi \left( a_3 ~ b_2 \right) \left( a_2 ~ b_1 \right) = \left( a_1 ~ a_2 \right) \left( b_1 ~ a_3 \right) \left( b_2 ~ b_3 \right) \left( a_4 ~ b_4 \right) \cdots$를 얻을 수 있으며, 따라서 $\pi\pi' = \left( a_1 ~ b_2 ~ a_3 \right) \left( a_2 ~ b_1 ~ b_3 \right) \in N$임을 알 수 있고, 따라서 &lt;b&gt;Case 2&lt;/b&gt;의 상황으로 환원된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Alternating Group</category>
      <category>Symmetric Group</category>
      <category>교대군</category>
      <category>대칭군</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/203</guid>
      <comments>https://chocobear.tistory.com/203#entry203comment</comments>
      <pubDate>Mon, 31 Jan 2022 12:59:25 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물두 번째 이야기  |  안정자군과 궤도 ( Stabilizer Subgroup &amp;amp; Orbit )</title>
      <link>https://chocobear.tistory.com/202</link>
      <description>&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Following the &lt;a href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;last post&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;, this post will explain the stabilizer subgroup. Let's look at the definition below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;&lt;a href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;저번 글&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;에 이어서, 이번 글에서는 안정자군에 대해 설명할 것이다. 아래의 정의를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;b&gt;Definition 1. Stabilizer Subgroup&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a group and $S$ be a &lt;a href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;$G$-set&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;. Let $s \in S$. The set of elements $x \in G$ such that $xs = s$ is obviously a subgroup of $G$, called the &lt;b&gt;stabilizer subgroup&lt;/b&gt; of $G$ with respect to $s$, or &lt;b&gt;isotropy group&lt;/b&gt; of $s$ in $G$, and denoted by $G_s$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$와 &lt;a href=&quot;https://chocobear.tistory.com/201&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;$G$-Set&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $S$가 주어졌으며, $s \in S$라고 하자. 그러면 $xs = s$가 되도록 하는 $x \in G$의 집합이 $G$의 부분군이 됨은 매우 당연하며, 이 부분군을 $G$의 $s$에 대한 &lt;b&gt;안정자군 ( Stabilizer Subgroup )&lt;/b&gt; 또는 $G$에서의 $s$의 &lt;b&gt;Isotropy Group&lt;/b&gt;이라고 부르며, $G_s$와 같이 쓴다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;When $G$ acts on itself by conjugation, then the isotropy group of an element is none other than the &lt;a href=&quot;https://chocobear.tistory.com/198&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;normalizer&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of this element. Similarly, when $G$ acts on the set of subgroups by conjugation, the isotropy group of a subgroup is again its normalizer.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we know what stabilizer subgroup is, let's find out the properties of stabilizer subgroup. Look at the theorems below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$가 자기자신 위로 켤레로서 작용할 때, 어떤 원소에 대한 안정자군은 다름 아닌 그 원소의 &lt;a href=&quot;https://chocobear.tistory.com/198&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Normalizer&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이며, 이와 비슷하게, $G$가 부분군의 집합 위로 켤레로서 작용하는 경우에는 어떤 부분군에 대한 안정자군은 그 부분군의 normalizer이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 안정자군이 무엇인지 알았으니, 안정자군의 성질들에 대해 알아보자. 아래의 정리들을 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let a group $G$ act on a set $S$. Let $s,s'$ be elements of $S$ and $y$ an element of $G$ such that $ys = s'$. Then $G_{s'} = yG_sy^{-1}$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$와 $G$-set $S$가 주어졌다고 하자. 만약 $s$와 $s'$이 $S$의 원소이고 $y \in G$에 대하여 $ys = s'$이 성립한다면, $G_{s'} = yG_sy^{-1}$이 성립한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $x$ be an element of $G$ such that $xs' = s'$. Then $xys = ys \Leftrightarrow y^{-1}xys = s \Leftrightarrow y^{-1}xy \in G_s \Leftrightarrow x \in yG_sy^{-1}$. Therefore, $G_{s'}$ is equal to $yG_sy^{-1}$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$x$를 $xs' = s'$이 성립하도록 하는 $G$의 원소라고 하자.&amp;nbsp; 그러면 $xys = ys \Leftrightarrow y^{-1}xys = s \Leftrightarrow y^{-1}xy \in G_s \Leftrightarrow x \in yG_sy^{-1}$가 성립하므로 $G_{s'}$은 $yG_sy^{-1}$이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 2.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $K$ be the &lt;a href=&quot;https://chocobear.tistory.com/141&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;kernel&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of the action $G \to \operatorname{Perm}(S)$ of $G$ on $S$. Then directly from the definitions, we obtain that $K = \displaystyle \bigcap_{s \in S} G_s$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$K$를 $S$ 위의 $G$의 action $G \to \operatorname{Perm}(S)$의 &lt;a href=&quot;https://chocobear.tistory.com/141&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Kernel&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이라고 하자. 그러면 정의로부터 $K = \displaystyle \bigcap_{s \in S} G_s$이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $s \in S$ and $x \in K$. Since $K$ is the kernel of the action $G \to \operatorname{Perm}(S)$, $xs = s$. Hence, $x \in G_s$ for all $s \in S$. Therefore, $x \in \displaystyle \bigcap_{s \in S} G_s$ and, hence, $K \subseteq \displaystyle \bigcap_{s \in S} G_s$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $x \in \displaystyle \bigcap_{s \in S} G_s$. Then $xs = s$ for all $s \in S$. So, $x$ is an element of the kernel of the action $G \to \operatorname{Perm}(S)$. Therefore, $x \in K$ and, hence, $\displaystyle \bigcap_{s \in S} G_s \subseteq K$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Therefore, $K = \displaystyle \bigcap_{s \in S} G_s$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$s \in S$이고 $x \in K$라고 하자. 그러면 $K$가 action $G \to \operatorname{Perm}(S)$의 kernel이라는 것으로부터 $xs = s$임을 알 수 있다. 따라서 임의의 $s \in S$에 대하여 $x \in G_s$가 성립함을 알 수 있으며, 따라서 $K \subseteq \displaystyle \bigcap_{s \in S} G_s$이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $x \in \displaystyle \bigcap_{s \in S} G_s$라고 하자. 그러면 임의의 $s \in S$에 대하여 $xs = s$가 성립하므로 $x$는 action $G \to \operatorname{Perm}(S)$의 kernel의 원소가 됨을 알 수 있다. 따라서 $\displaystyle \bigcap_{s \in S} G_s \subseteq K$이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위 두 사실을 종합하면 $K = \displaystyle \bigcap_{s \in S} G_s$임을 알 수 있다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition2&quot;&gt;&lt;/a&gt;&lt;b&gt;Definition 2. Orbit&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let a group $G$ act on a set $S$ and let $s \in S$. The subset of $S$ consisting of all elements $xs$ with $x \in G$ is denoted by $Gs$, and is called the &lt;b&gt;orbit&lt;/b&gt; of $s$ under $G$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$가 집합 $S$ 위로 작용한다고 하고 $s \in S$라고 하자. 그러면 $S$의 부분집합 $\{ xs \;|\; x \in G \}$을 $Gs$와 같이 나타내고, 이를 $G$ 아래에서의 $s$의 &lt;b&gt;궤도 ( Orbit )&lt;/b&gt;라고 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem3&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 3.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;If $G$ is a group acting on a set $S$ and $s \in S$, then the order of the orbit $Gs$ is equal to the &lt;a href=&quot;https://chocobear.tistory.com/140&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;index&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $\left( G : G_s \right)$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;만약 $G$가 집합 $S$ 위로 작용하는 군이고 $s \in S$라면, 궤도 $Gs$의 위수는 &lt;a href=&quot;https://chocobear.tistory.com/140&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;지표&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $\left( G : G_s \right)$와 같다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since it is obvious, leave the proof to the readers.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;비교적 자명하므로, 증명은 독자에게 맡긴다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem4&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 4.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;The number of conjugate subgroups to $H$ is equal to the index of the normalizer of $H$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$H$와 켤레관계인 부분군의 개수는 $H$의 normalizer의 지표와 같다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Let $a,b \in G$ and $N_H$ be the normalizer of $H$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Then $a^{-1}Ha = b^{-1}Hb \Leftrightarrow (ab^{-1})^{-1}Hab^{-1} = H \Leftrightarrow ab^{-1} \in N_H \Leftrightarrow a \in N_Hb$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;So, two distinct elements of $G$ will produce the same conjugate of $H$ if and only if they belong to the same right coset of $N_H$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Thus, the number of conjugate subgroups to $H$ is equal to the number of right cosets of $N_H$, which is the index of the normalizer of $H$. Therefore, $\lvert \{ a^{-1}Ha \mid a \in G \} \rvert = \left( G : N_H \right)$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$a,b \in G$라고 하고 $N_H$를 $H$의 normalizer라고 하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러면 $a^{-1}Ha = b^{-1}Hb \Leftrightarrow (ab^{-1})^{-1}Hab^{-1} = H \Leftrightarrow ab^{-1} \in N_H \Leftrightarrow a \in N_Hb$이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러므로 $G$의 서로 다른 두 원소가 생성하는 $H$의 켤레가 같은 것은 그 두 원소가 $N_H$의 같은 우잉여류에 속하는 것과 필요충분조건이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 $H$와 켤레관계인 부분군의 개수와 $N_H$의 우잉여류의 개수는 같으며, $N_H$의 우잉여류의 개수가 곧 $N_H$의 지표이므로 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Corollary 4.1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a group and $H$ a subgroup of index $2$. Then $H$ is normal in $G$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;$H$를 군 $G$의 지표가 $2$인 부분군이라고 하자. 그러면 $H$는 정규부분군이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Note that $H$ is contained in its normalizer $N_H$, so the index of $N_H$ is $1$ or $2$. If the index of $N_H$ is $1$, then we are done.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Suppose it is $2$. Let $G$ act by conjugation on the set of subgroups. The orbit of $H$ has two elements since the &lt;a href=&quot;#Theorem4&quot;&gt;&lt;b&gt;Theorem 4&lt;/b&gt;&lt;/a&gt;, and $G$ acts on this orbit. In this way we get a homomorphism of $G$ into the group of permutations of $2$ elements. Since there is one conjugate of $H$ unequal to $H$, then the kernel of our homomorphism is normal, of index $2$, hence equal to $H$, which is normal, a contradiction which concludes the proof.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$H$가 $H$의 normalizer $N_H$에 포함된다는 사실을 기억해두자. 따라서 $N_H$의 지표는 $1$ 또는 $2$이다. 만약 $N_H$의 지표가 $1$이라면, 증명이 완료된다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 $N_H$의 지표가 $2$라고 가정하자. 이제 $G$가 그의 부분군의 집합 위로 켤레로서 작용한다고 하자. 그러면&amp;nbsp;&lt;a href=&quot;#Theorem4&quot;&gt;&lt;b&gt;Theorem 4&lt;/b&gt;&lt;/a&gt; $H$의 궤도는 두 개의 원소를 가지며, $G$는 이 궤도 위로 작용한다. 이렇게 하면 $G$에서 두 원소의 순열의 군으로의 homomorphism을 얻을 수 있다. $H$와 다른 $H$의 켤레가 단 하나 존재한다는 사실로부터 앞서 얻어낸 homomorphism의 kernel이 정규부분군이며, 그의 지표가 $2$임을 알 수 있고, 따라서 그 kernel은 $H$와 같음을 알 수 있다. 따라서 정리가 증명된다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>orbit</category>
      <category>Stabilizer Subgroup</category>
      <category>궤도</category>
      <category>안정자군</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/202</guid>
      <comments>https://chocobear.tistory.com/202#entry202comment</comments>
      <pubDate>Wed, 26 Jan 2022 06:39:24 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스물한 번째 이야기  |  군의 작용 ( Group Action )</title>
      <link>https://chocobear.tistory.com/201</link>
      <description>&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, we will learn about the group action. In mathematics, a group action on a set is a group homomorphism of a given group into the group of permutations of the set. Let's look at the definition below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 group action에 대해 알아볼 것이다. 집합 $S$ 위의 group action은 주어진 군에서 $S$ 위의 순열의 군으로 가는 homomorphism을 말한다. 아래의 정의를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Let $G$ be a group and let $S$ be a set. An &lt;b&gt;action&lt;/b&gt; of $G$ on $S$ is a homomorphism $\pi : G \to \operatorname{Perm}(S)$ of $G$ into the group of permutations of $S$. We then call $S$ a &lt;b&gt;$G$-set&lt;/b&gt;. We denote the permutation associated with an element $x \in G$ by $\pi_x$ and thus the homomorphism is denoted by $x \mapsto \pi_x$. We often abbreviate the notation and write simply $xs$ instead of $\pi_x(s)$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$와 집합 $S$가 주어졌다고 하자. $S$ 위의 $G$의 &lt;b&gt;작용 ( Action )&lt;/b&gt;은 $G$에서 $S$의 순열의 군 $\operatorname{Perm}(S)$로 가는 homomorphism $\pi : G \to \operatorname{Perm}(S)$이다. 이러한 action이 주어진 경우, $S$를 &lt;b&gt;$G$-Set&lt;/b&gt;이라고 부르며, $x \in G$가 대응되는 순열을 $\pi_x$와 같이 나타내며, 따라서 앞서 언급한 homomorphism은 $x \mapsto \pi_x$로 나타낸다. 일반적으로, group action의 경우, $\pi_x (s)$ 대신에 $xs$와 같이 나타낸다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;With the simpoler notation above, we have the two familiar properties. Look at the theorem below:&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위의 표기 중 더 간단한 표기를 사용한다면, 두 가지 익숙한 성질을 얻을 수 있다. 아래의 정리를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 1.&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;The following statements are satisfied:&lt;br /&gt;(a) For all $x,y \in G$ and $s \in S$, we have $x(ys) = (xy)s$.&lt;br /&gt;(b) If $e$ is the identity element of $G$, then $es = s$ for all $s \in S$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;아래의 명제들이 성립한다.&lt;br /&gt;(a) 임의의 $x,y \in G$와 $s \in S$에 대하여 $x(ys) = (xy)s$가 성립한다.&lt;br /&gt;(b) 만약 $e$가 $G$의 항등원이라면, 임의의 $s \in S$에 대하여 $es = s$가 성립한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (a)&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $x \mapsto \pi_x$ is a group-homomorphism, $x(ys) = \pi_x ( \pi_y (s) ) = ( \pi_x \circ \pi_y ) (s) = \pi_{xy} (s) = (xy)s$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$x \mapsto \pi_x$가 group-homomorphism이므로 $x(ys) = \pi_x ( \pi_y (s) ) = ( \pi_x \circ \pi_y ) (s) = \pi_{xy} (s) = (xy)s$가 성립한다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (b)&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since $x \mapsto \pi_x$ is a group-homomorphism, $\pi_e$ must be an identity element of the group $\operatorname{Perm}(S)$. Thus, $\pi_e$ is the identity map on $S$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$x \mapsto \pi_x$가 group-homomorphism이므로 $\pi_e$는 군 $\operatorname{Perm}(S)$의 항등원이어야 한다. 따라서 $\pi_e$는 $S$ 위의 항등함수이다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Conversely, if we are given a mapping $G \times S \to S$, denoted by $( x , s ) \mapsto xs$, satisfying these two properties, then for each $x \in G$ the map $s \mapsto xs$ is a permuation of $S$, which we then denote by $\pi_x(s)$. Then $x \mapsto \pi_x$ is a homomorphism of $G$ into $\operatorname{Perm}(S)$. So an action of $G$ on $S$ could also be defined as a mapping $G \times S \to S$ satisfying the above two properties. The most important examples of representations of $G$ as a group of permutations are the following.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;역으로, 만약&amp;nbsp;&lt;a href=&quot;#Theorem1&quot;&gt;&lt;b&gt;Theorem 1&lt;/b&gt;&lt;/a&gt;의 두 명제 (a), (b)를 만족하는 사상 $G \times S \to S : ( x , s ) \mapsto xs$가 주어진다면, 각 $x \in G$에 대하여 사상 $s \mapsto xs$는 $S$ 위의 순열임을 알 수 있으며, 이 사상을 $\pi_x$로 나타내게 되면, $x \mapsto \pi_x$는 $G$에서 $\operatorname{Perm}(S)$로 가는 group-homomorphism이 됨을 알 수 있다. 따라서 $S$ 위의 $G$의 action은 위의 두 성질을 만족하는 사상 $G \times S \to S$로 정의할 수도 있다. 아래는 이러한 $G$의 action의 몇 가지 예시이다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Example 1. Conjugation&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;For each $x \in G$, let $\operatorname{c}_x : G \to G$ be the map such that $\operatorname{c}_x (y) = xyx^{-1}$. Then it is immediately verified that the association $x \mapsto \operatorname{c}_x$ is a homomorphism $G \to \operatorname{Aut}(G)$, and so this map gives an action of $G$ on itself, called conjugation. The kernel of the homomorphism $x \mapsto \operatorname{c}_x$ is a normal subgroup of $G$, which consists of all $x \in G$ such that $xyx^{-1} = y$ for all $y \in G$, i.e., all $x \in G$ which commute with every element of $G$. This kernel is called the &lt;a href=&quot;https://chocobear.tistory.com/198&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;center&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of $G$. &lt;a href=&quot;https://chocobear.tistory.com/118&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;automorphisms&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of $G$ of the form $\operatorname{c}_x$ are called &lt;b&gt;inner&lt;/b&gt;.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;To avoid confusion about the action on the left, we don't write $xy$ for $\operatorname{c}_x (y)$. Sometimes, one writes $\operatorname{c}_{x^{-1}} (y) = x^{-1}yx = y^x$, i.e., one uses exponential notation, so that we have the rules $y^{xz} = (y^x)^z$ and $y^e = y$ for all $x,y,z \in G$. Similarly, ${}^xy = xyx^{-1}$ and ${}^z(^xy) = {}^{zx}y$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;We note that $G$ also acts by conjugation on the set of subsets of $G$. Indeed, let $S$ be the set of subsets of $G$, and let $A \in S$ be a subset of $G$. Then $xAx^{-1}$ is also a subset of $G$ which may be denoted by $\operatorname{c}_x (A)$, and one verifies trivially that the map $(x,A) \mapsto xAx^{-1}$ of $G \times S \to S$ is an action of $G$ on $S$. We note in addition that if $A$ is a subgroup of $G$ then $xAx^{-1}$ is also a subgroup, so that $G$ acts on the set of subgroups by conjugation.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;각 $x \in G$에 대하여 $\operatorname{c}_x : G \to G$를 $\operatorname{c}_x (y) = xyx^{-1}$과 같이 정의되는 함수라고 하자. 그러면 $x \mapsto \operatorname{c}_x$가 $G$에서 $\operatorname{Aut}(G)$로 가는 group-homomorphism이 됨은 자명하며, 이 사상이 켤레라고 불리는, $G$ 위의 $G$의 action을 정의함을 알 수 있다. 이때, $x \mapsto \operatorname{c}_x$의 kernel은 $G$의 정규부분군이 되며, 이를 $G$의 &lt;a href=&quot;https://chocobear.tistory.com/198&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Center&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;라고 한다. 또한, $G$의 &lt;a href=&quot;https://chocobear.tistory.com/118&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Automorphism&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; 중 $\operatorname{c}_x$ 꼴의 함수들을&amp;nbsp;&lt;b&gt;Inner&lt;/b&gt;라고 한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;왼쪽에서 작용하는 것과 혼동하지 않게 하기 위해 $\operatorname{c}_x (y)$는 $xy$로 나타내지 않는다. 대신 일부 저자들은 $\operatorname{c}_{x^{-1}} (y) = x^{-1}yx = y^x$와 같이 나타내기도 하며, 비슷하게 $\operatorname{c}_x ( y )$는 ${}^x y$와 같이 나타낸다. 당연하지만, 임의의 $x,y,z \in G$에 대하여 $y^{xz} = (y^x)^z$와 $y^e = y$가 성립한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$G$는 $G$ 위로 작용할 뿐 아니라 $G$의 멱집합 위로도 작용한다. $S$를 $G$의 모든 부분집합의 집합이라 하고 $A \in S$를 생각하자. 그러면 $xAx^{-1}$ 역시 $G$의 부분집합이 되며, 이를 $\operatorname{c}_x (A)$와 같이 나타낼 수 있다. 그러면 자명하게도 $G \times S \to S : (x,A) \mapsto xAx^{-1}$이 $S$ 위의 $G$의 action이 되며, 만약 $A$가 $G$의 부분군이라면, $xAx^{-1}$ 역시 $G$의 부분군이 되므로 $G$는 그의 부분군의 집합 위로 켤레로서 작용함을 알 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Example 2. Translation&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;For each $x \in G$ we define the &lt;b&gt;translation&lt;/b&gt; $\operatorname{T}_x : G \to G$ by $\operatorname{T}_x (y) = xy$. Then the map $(x,y) \mapsto xy = \operatorname{T}_x(y)$ defines an action of $G$ on itself.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Similarly, $G$ acts by translation on the set of subsets, for if $A$ is a subset of $G$, then $xA = \operatorname{T}_x (A)$ is also subset of $G$. If $H$ is a subgroup of $G$, then $\operatorname{T}_x (H) = xH$ is in general not a subgroup but a coset of $H$, and hence we see that $G$ acts by translation on the set of cosets of $H$. We denote the set of left cosets of $H$ by $G/H$. Thus even though $H$ need not be normal, $G/H$ is a $G$-set. It has become customary to denote the set of right cosets by $G \backslash H$.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;각 $x \in G$에 대하여 $\operatorname{T}_x : G \to G$를 $\operatorname{T}_x (y) = xy$와 같이 정의되는 함수라고 하자. 그러면 $(x,y) \mapsto xy = \operatorname{T}_x(y)$는 $G$ 위의 $G$의 action을 정의한다. 이를 &lt;b&gt;Translation&lt;/b&gt;이라고 한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;비슷하게, $G$는 $G$의 멱집합 위로 translation으로서 작용한다. 만약 $A$가 $G$의 부분집합이라면, $xA = \operatorname{T}_x (A)$ 역시 $G$의 부분집합이다. 또한, 만약 $H$가 $G$의 부분군이라면, $\operatorname{T}_x (H) = xH$는 $H$의 부분군이 아닐 수는 있지만, $H$의 coset이 되며, 따라서 $G$는 $H$의 coset의 집합 위로 translation으로서 작용함을 알 수 있다. 따라서 $H$가 $G$의 정규부분군이 아니라 하더라도 $G/H$가 $G$-set임을 알 수 있다.&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Group Action</category>
      <category>군의 작용</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/201</guid>
      <comments>https://chocobear.tistory.com/201#entry201comment</comments>
      <pubDate>Wed, 26 Jan 2022 05:05:03 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 스무 번째 이야기 | Semidirect Product &amp;amp; Direct Product</title>
      <link>https://chocobear.tistory.com/200</link>
      <description>&lt;p data-ke-size=&quot;size16&quot;&gt;이번 글에서는 직접곱과 반직접곱에 대한 이야기를 해보려 한다. 역시 선형대수학을 공부해본 독자라면 아마 두 벡터공간의 direct product 혹은 direct sum에 대하여 접해보았을 것이다. 그러나 군에서는 그 정의가 살짝 다르다. 아래의 정의를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 1. Direct Product (Finite)&lt;br /&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;군 $G_1,G_2, \cdots G_n$에 대하여 다음과 같이 정의되는 $G = G_1 \times G_2 \times \cdots \times G_n$을 군 $G_1,G_2, \cdots G_n$의 &lt;b&gt;Direct Product&lt;/b&gt;라고 한다. &lt;br /&gt;&lt;br /&gt;(a) 군의 집합으로 $G_1,G_2, \cdots , G_n$의 &lt;a href=&quot;https://chocobear.tistory.com/13&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Cartesian Product&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;를 가진다. 즉,&amp;nbsp; $G$는 모든 $i = 1,2, \cdots ,n$에 대해 $g_i \in G_i$를 만족하는 $n$-tuple들의 집합이다. &lt;br /&gt;&lt;br /&gt;(b) 연산은 다음과 같이 정의한다.&lt;br /&gt;각 $G_i$의 연산을 $*_i$ 라고 하고, $G$의 연산을 $*$라고 하자. $$(g_1,g_2, \cdots, g_n) * (h_1,h_2,\cdots, h_n) = (g_1 *_1 h_1, g_2 *_2 h_2, \cdots, g_n *_n h_n)$$그러면 $G$는 군의 공리를 모두 만족한다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;하지만 위와 같이 $n$-tuple을 사용하는 정의는 개수가 유한할 때만 사용할 수 있다는 제한이 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;서적에 따라서 함수를 사용하여 이러한 문제를 해결한 정의를 이용하는 경우도 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 2. Direct Product / Sum&lt;br /&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;군의 집합 $G_{i_1},G_{i_2}, \cdots$에 대하여 다음과 같이 정의되는 $G = G_{i_1} \times G_{i_2} \times \cdots $을 군 $G_{i_1},G_{i_2}, \cdots $의 &lt;b&gt;Direct Product&lt;/b&gt;라고 한다. &lt;br /&gt;index set $I = \{ i_1,i_2, \cdots \}$를 이용하여 $\{G_i\}_{i \in I}$의 direct product라고 하기도 한다. &lt;br /&gt;&lt;br /&gt;(a) 군의 집합으로 $G_1,G_2, \cdots , G_n$의 cartesian product를 가진다. 즉 $G$를 다음과 같이 함수들의 집합으로 정의한다. $$ G = \{ f : I \to \displaystyle\bigcup_{i \in I} G_i \;|\; \forall i \in I, f(i) \in G_i \} $$&lt;br /&gt;(b) 연산은 다음과 같이 정의한다.&lt;br /&gt;각 $G_i$의 연산, 항등원을 각각 $*_i,1_i$ , $G$의 연산을 $*$라고 하자. $$\forall i \in I, (f * g)(i) = f(i) *_i g(i)$$&lt;br /&gt;그러면 $G$는 군의 공리를 모두 만족한다.&lt;br /&gt;또, $$ G = \{ f : I \to \displaystyle\bigcup_{i \in I} G_i \mid \lvert \{ i \in I \mid f(i) \neq 1_i \} \rvert &amp;lt; \infty \land \forall i \in I, f(i) \in G_i \} $$ 로 수정하면 이를 &lt;b&gt;Direct Sum&lt;/b&gt;이라고 한다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;이제 direct product의 정의에 대해 알았으니, semidirect product에 대하여 알아보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Definition 3. Semidirect Product &lt;br /&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%; height: 19px;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr style=&quot;height: 19px;&quot;&gt;
&lt;td style=&quot;width: 100%; height: 19px;&quot;&gt;군 $H,K$, homomorphism $\phi : K \to \operatorname{Aut}(H)$가 주어졌을 때, $G = H \rtimes_{\phi} K$를 $N$과 $H$의 $\phi$를 통한 &lt;b&gt;Semidirect Product&lt;/b&gt;라고 한다. 이때, $\operatorname{Aut}(H)$는 $H$ 위의 &lt;a href=&quot;https://chocobear.tistory.com/118&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Automorphism&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;의 집합이다.&lt;br /&gt;&lt;br /&gt;(a) 군의 집합으로 $H \times K$를 가진다. &lt;br /&gt;&lt;br /&gt;(b) 연산은 다음과 같이 정의한다. &lt;br /&gt;$H,K,G$의 연산을 각각 $*_H,*_K,*_G$라고 하자. $(h_1,k_1) *_G (h_2,k_2) = (h_1 *_H (\phi(k_1))(h_2), k_1 *_K k_2 )$&lt;br /&gt;&lt;br /&gt;그러면 $G$는 군의 공리를 모두 만족한다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;주로 $\phi(k)(h) = khk^{-1}$와 같이 &lt;a href=&quot;https://chocobear.tistory.com/128&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;left conjugation&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; action으로 정의하는 경우가 잦고, 이러한 경우 $\phi$를 생략하여 $H \rtimes K$라고 쓰기도 한다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;먼저, semidirect product에 관련된 몇 가지 동치 명제들을 알아보고, 그 뒤 연산에 부여되는 homomorphism이 left conjugation action일때 새로이 얻게 되는 동치 명제들에 대하여 알아보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 1. Equivalent Propositions&lt;br /&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%; height: 19px;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr style=&quot;height: 19px;&quot;&gt;
&lt;td style=&quot;width: 100%; height: 19px;&quot;&gt;군 $H,K$, &lt;a href=&quot;https://chocobear.tistory.com/118&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;Homomorphism&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; $\phi : K \to Aut(H)$가 주어졌을 때 아래 세 명제는 서로 동치이다. &lt;br /&gt;&lt;br /&gt;(a) $H \rtimes_{\phi} K$ 와 $H \times K$ 사이 identity map (항등 사상)은 group-homomorphism이 된다. &lt;br /&gt;&lt;br /&gt;(b) $\phi = \operatorname{id}_H$이다. ( $\operatorname{id}_H$는 $H \to H$로 가는 identity map이다. )&lt;br /&gt;&lt;br /&gt;(c) $K' = \{ (1,k') | k' \in K \} \trianglelefteq G = H \rtimes_{\phi} K $이다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (a) -&amp;gt; (b)&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;(a)의 가정에 의하여 $$ \forall h_2 \in H, \forall k_1 \in K, \phi(k_1)(h_2) = h_2 $$&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (b) -&amp;gt; (c)&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;임의의 $g = (h,k) \in G$, $h = (1,k') \in K'$에 대하여 $ghg^{-1} \in K'$를 보이자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;먼저, (b)의 가정에 의하여, $(h,k)^{-1} = (h^{-1},k^{-1})$이다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;$$(h,k)(1,k')(h,k)^{-1} = (h,kk')(h,k)^{-1} = (1,kk'k^{-1}) \in K'$$&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (c) -&amp;gt; (a)&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;먼저, $H' = \{ (h',1) | h' \in H \} \trianglelefteq G = H \rtimes_{\phi} K $ 임을 앞선 논의와 유사하게 얻을 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;따라서 임의의 $h \in H' , k \in K'$에 대해 &lt;b&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;a style=&quot;color: #006dd7;&quot; href=&quot;https://chocobear.tistory.com/195?category=912759&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;교환자&lt;/a&gt;&lt;/span&gt;&lt;/b&gt; $[h,k]$를 생각하면 $[h,k] \in H \cap K = 1 $이 되어, $H',K'$가 가환임을 얻는다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;또, $\phi$의 함수값은 반드시 homomorphism이 되어야 하므로, 항등원이 항등원으로 대응된다는 사실을 이용할 수 있고, 또한 역원에 대해서도 정보를 얻을 수 있다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;이를 이용하여 다음과 같은 식을 생각하자. $$\begin{align*}(\phi(k)(h),1) &amp;amp;= (\phi(k)(h) \phi(k)(1), kk^{-1}) = (\phi(k)(h),k)(1,k^{-1}) \\ &amp;amp;= (1,k) (h,1)(1,k^{-1}) = (1,k)(h,1)(1,k)^{-1} = (h,1)\end{align*}$$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;위 식은 임의의 $h \in H' , k \in K'$에 대해 성립하므로, $\phi$가 항등사상이라는 사실을 얻고, 자연스레 (a)를 얻는다.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;이제 $\phi$가 left conjugation action으로 정의될 때 성립하는 동치인 명제들에 대해 알아보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Theorem 2. Equivalent Propositions&lt;br /&gt;&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%; height: 19px;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr style=&quot;height: 19px;&quot;&gt;
&lt;td style=&quot;width: 100%; height: 19px;&quot;&gt;군 $G$와 이의 항등원 $1$, $H \leq G , N \trianglelefteq G$라고 하자. 그러면 다음 네 명제는 서로 동치이다. &lt;br /&gt;&lt;br /&gt;(a) $G = NH, N \cap H = 1$&lt;br /&gt;&lt;br /&gt;(b) $\forall g \in G, \exists ! n \in N, h \in H : g = nh$&lt;br /&gt;&lt;br /&gt;(c) $\forall g \in G, \exists ! n \in N, h \in H : g = hn$&lt;br /&gt;&lt;br /&gt;(d) $\iota : \begin{cases} H \to G \\ h \mapsto h \end{cases}$ , $\pi : \begin{cases} G \to G/N \\ g \mapsto gN \end{cases}$에 대해, $\pi \circ \iota$는 isomorphism이 된다.&lt;br /&gt;&lt;br /&gt;그리고 이러한 경우 $G = N \rtimes H$라고 한다.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;증명은 어렵지 않으므로 생략한다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;또한, 한 가지 주의할 점은 $\phi$에 따라 $G = H \rtimes_{\phi} K $가 달라질 수 있고, 서로 isomorphic하지 않을 수 있다는 것이다.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;예를 들어, 위수가 $8$인 순환군과 위수가 $2$인 순환군의 direct product는 이들의 semidirect product이며, 또한, 위수가 $16$인 dihedral group 역시 위수가 $8$인 순환군과 위수가 $2$인 순환군의 semidirect product이다. 그러나, 위수가 $16$인 dihedral group은 &lt;a href=&quot;https://chocobear.tistory.com/126&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;아벨군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이 아니므로 위수가 $8$인 순환군과 위수가 $2$인 순환군의 direct product와 isomorphic하지 않다.&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Direct Product</category>
      <category>Semidirect Product</category>
      <category>반직접곱</category>
      <category>직접곱</category>
      <category>추상대수학</category>
      <author>SuminKim</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/200</guid>
      <comments>https://chocobear.tistory.com/200#entry200comment</comments>
      <pubDate>Sun, 23 Jan 2022 01:37:56 +0900</pubDate>
    </item>
    <item>
      <title>추상대수학, 그 열아홉 번째 이야기  |  군의 부분집합의 Centralizer와 Normalizer ( Centralizer &amp;amp; Normalizer of Subset in Group )</title>
      <link>https://chocobear.tistory.com/198</link>
      <description>&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이번 글에서는 군의 부분집합의 centralizer와 normalizer에 대해 알아볼 것이다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;먼저, centralizer가 무엇인지 알아보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In this post, we will learn about the centralizer and normalizer of a subset in a group.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;First of all, let's find out what centralizer is.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition1&quot;&gt;&lt;/a&gt;&lt;b&gt;Definition 1. Centralizer&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;The &lt;b&gt;centralizer&lt;/b&gt; $\mathrm{C}_G (S)$ of a subset $S$ in a group $G$ is defined as $$ \mathrm{C}_G (S) = \{ g \in G \;|\; \forall s \in S ,\; gs = sg \} = \{ g \in G \;|\; \forall s \in S ,\; gsg^{-1} = s \}. $$ If there is no ambiguity about the group in question, the $G$ can be suppressed from the notation. When $S = \{ a \}$ is a singleton set, we write $\mathrm{C}_G (a)$ instead of $\mathrm{C}_G(\{a\})$.&lt;br /&gt;Furthermore, the centralizer of a group $G$ itself is called the &lt;b&gt;center&lt;/b&gt; of a group $G$. The center of a group $G$ is denoted by $\mathrm{Z}(G)$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$의 부분집합 $S$의&amp;nbsp;&lt;b&gt;Centralizer&lt;span&gt;&amp;nbsp;&lt;/span&gt;&lt;/b&gt;$\mathrm{C}_G (S)$는 다음과 같이 정의된다. $$ \mathrm{C}_G (S) = \{ g \in G \;|\; \forall s \in S ,\; gs = sg \} = \{ g \in G \;|\; \forall s \in S ,\; gsg^{-1} = s \} $$ 만약 맥락상 어떤 군에서의 centralizer를 말하는 것인지에 대한 모호함이 없는 경우라면, $\mathrm{C}_G (S)$를 $\mathrm{C} (S)$와 같이 나타내기도 한다. 만약 $S = \{ a \}$가 단원소 집합이라면, $\mathrm{C}_G (\{ a \})$라고 쓰는 대신 $\mathrm{C}_G (a)$라고 쓰기도 한다.&lt;br /&gt;또한, 군 $G$의 centralizer는 $G$의&amp;nbsp;&lt;b&gt;Center&lt;/b&gt;라고 부르며, $\mathrm{Z}(G)$와 같이 나타낸다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 centralizer가 무엇인지 알았으니 normalizer가 무엇인지 알아보자.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we know what centralizer is, let's find out what normalizer is.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Definition2&quot;&gt;&lt;/a&gt;&lt;b&gt;Definition 2. Normalizer&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;The &lt;b&gt;normalizer &lt;/b&gt;$\mathrm{N}_G (S)$ of a subset $S$ in a group $G$ is defined as $$ \mathrm{N}_G (S) = \{ g \in G \;|\; gS = Sg \} = \{ g \in G \;|\; gSg^{-1} = S \}. $$ The same notational conventions mentioned above for centralizers also apply to normalizers.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;군 $G$의 부분집합 $S$의&amp;nbsp;&lt;b&gt;Normalizer&lt;/b&gt; $\mathrm{N}_G (S)$는 다음과 같이 정의된다. $$ \mathrm{N}_G (S) = \{ g \in G \;|\; gS = Sg \} = \{ g \in G \;|\; gSg^{-1} = S \} $$ centralizer에서와 같이, 만약 맥락상 어떤 군에서의 normalizer를 말하는 것인지에 대한 모호함이 없는 경우라면, $\mathrm{N}_G (S)$를 $\mathrm{N} (S)$와 같이 나타내기도 하며, $S = \{ a \}$가 단원소 집합이라면, $\mathrm{N}_G (\{ a \})$라고 쓰는 대신 $\mathrm{N}_G (a)$라고 쓰기도 한다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;The definitions are similar but not identical. If $g$ is in the centralizer of $S$ and $s$ is in $S$, then it must be that $gs = sg$, but if $g$ is in the normalizer, then $gs = tg$ for some $t$ in $S$, with $t$ possibly different from $s$. That is, elements of the centralizer of $S$ must commute pointwise with $S$, but elements of the normalizer of $S$ need only commute with $S$ as a set.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now that we have defined centralizer and normalizer, it is time to think about the properties of them. Let's look at the theorem below.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;위 두 정의는 비슷해보이지만, 완전히 동일하지는 않다. 만약 $g$가 $S$의 centralizer의 원소이고, $s \in S$라면, $gs = sg$이어야 하지만, $g$가 normalizer의 원소라면, 어떤 $t \in S$에 대하여 $gs = tg$가 성립하는 것으로도 충분하다. 즉, $S$의 centralizer의 원소는 $S$의 각 원소들과 commute하지만, $S$의 normalizer의 원소는 $S$의 각 원소들과 commute할 필요는 없으며, 집합 $S$ 그 자체와 commute하기만 해도 충분하다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;centralizer와 normalizer를 정의했으니 이제 이들의 성질에 대해 알아볼 차례이다. 아래의 정리를 보자.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;a name=&quot;Theorem1&quot;&gt;&lt;/a&gt;&lt;b&gt;Theorem 1. Properties of Centralizers and Normalizers&lt;/b&gt;&lt;/p&gt;
&lt;table style=&quot;border-collapse: collapse; width: 100%;&quot; border=&quot;1&quot; data-ke-align=&quot;alignLeft&quot;&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td style=&quot;width: 100%;&quot;&gt;
&lt;div class=&quot;en&quot;&gt;Belows are the properties of the centralizers and normalizers:&lt;br /&gt;(a) The centralizer and normalizer of subset $S$ in a group $G$ are both subgroups of $G$.&lt;br /&gt;(b) The centralizer of $S$ is always a &lt;a href=&quot;https://chocobear.tistory.com/128#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;normal subgroup&lt;/b&gt;&lt;/span&gt;&lt;/a&gt; of the normalizer of $S$.&lt;br /&gt;(c) $\mathrm{C}_G ( \mathrm{C}_G (S) )$ contains $S$.&lt;br /&gt;(d) If $H$ is a subgroup of $G$, then the normalizer of $H$ is the largest subgroup of $G$ in which $H$ is normal.&lt;br /&gt;(e) If $S$ is a subset of $G$ such that all elements of $S$ commute with each other, then the centralizer of $S$ is the largest subgroup of $G$ whose center contains $S$.&lt;/div&gt;
&lt;div class=&quot;ko&quot;&gt;아래는 centralizer와 normalizer의 성질들이다.&lt;br /&gt;(a) 군 $G$의 부분집합 $S$가 주어질 때, $S$의 centralizer와 normalizer는 $G$의 부분군이다.&lt;br /&gt;(b) $S$의 centralizer는 항상 $S$의 normalizer의 &lt;a href=&quot;https://chocobear.tistory.com/128#Definition2&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span style=&quot;color: #006dd7;&quot;&gt;&lt;b&gt;정규부분군&lt;/b&gt;&lt;/span&gt;&lt;/a&gt;이다.&lt;br /&gt;(c) $S \subseteq \mathrm{C}_G ( \mathrm{C}_G (S) )$.&lt;br /&gt;(d) $H$가 $G$의 부분군이라면, $H$의 normalizer는 $H$를 정규부분군으로 가지는 $G$의 가장 큰 부분군이다.&lt;br /&gt;(e) 군 $G$의 부분집합 $S$에 대하여 $S$의 원소 사이의 교환법칙이 성립한다면, $S$의 centralizer는 center가 $S$를 포함하는 $G$의 가장 큰 부분군이다.&lt;/div&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Proof.&lt;/b&gt;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (a), (b), (c)&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;자명하므로 증명은 독자에게 맡긴다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Since it is obvious, leave the proof to the readers.&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (d)&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$H$가 $\mathrm{N}_G (H)$의 정규부분군임은 자명하다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $H$를 정규부분군으로 가지는 $K \leq G$를 생각하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러면 임의의 $k \in K$에 대하여 $kHk^{-1} = H$가 성립한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 normalizer의 정의에 의해 $K$는 $H$의 normalizer의 부분집합이며, 따라서 $K \leq \mathrm{N}_G (H)$이다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;It is clear that $H$ is normal in the normalizer of $H$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now consider a subgroup $K$ of the group $G$ in which $H$ is normal.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Then $kHk^{-1} = H$ holds for every $k \in K$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Hence, by the definition of normalizer, $K$ is a subset of the normalizer of $H$ and $K$ is therefore a subgroup of $\mathrm{N}_G (H)$.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&lt;b&gt;Part (e)&lt;/b&gt;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;$S$의 원소 사이의 교환법칙이 성립한다면 $S$가 $\mathrm{Z} ( \mathrm{C}_G (S) )$의 부분집합임은 자명하다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이제 $K$의 center가 $S$를 포함하도록 하는 $K \leq G$를 생각하자.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;그러면 임의의 $k \in K$와 $s \in S$에 대하여 $sk = ks$가 성립한다.&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;따라서 centralizer의 정의에 의해 $K$는 $S$의 centralizer의 부분집합이며, 따라서 $K \leq \mathrm{C}_G (H)$이다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;If all elements of $S$ commute with each other, then it is obvious that the center of the centralizer of $S$ contains the set $S$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Now let $K$ be a subgroup of the group $G$ whose center contains $S$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In that condition, $sk = ks$ holds for every $k \in K$ and $s \in S$.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;Therefore, by the definition of centralizer, the centralizer of $S$ contains $K$ and, hence, $K$ is a subgroup of the centralizer of $H$.&lt;/p&gt;
&lt;p style=&quot;text-align: right;&quot; data-ke-size=&quot;size16&quot;&gt;$\blacksquare$&lt;/p&gt;
&lt;p data-ke-size=&quot;size16&quot;&gt;&amp;nbsp;&lt;/p&gt;
&lt;p class=&quot;ko&quot; data-ke-size=&quot;size16&quot;&gt;이 외에도 다양한 성질들이 더 있으나, 분량상 생략하도록 하겠다.&lt;/p&gt;
&lt;p class=&quot;en&quot; data-ke-size=&quot;size16&quot;&gt;In addition, there are various other properties, but they are omitted for concision.&lt;/p&gt;</description>
      <category>수학/추상대수학 | Abstract Algebra</category>
      <category>Abstract Algebra</category>
      <category>Center of Group</category>
      <category>Centralizer</category>
      <category>Normalizer</category>
      <category>군의 중심</category>
      <category>정규화 부분군</category>
      <category>중심화 부분군</category>
      <category>추상대수학</category>
      <author>초코맛 도비</author>
      <guid isPermaLink="true">https://chocobear.tistory.com/198</guid>
      <comments>https://chocobear.tistory.com/198#entry198comment</comments>
      <pubDate>Fri, 21 Jan 2022 15:09:54 +0900</pubDate>
    </item>
  </channel>
</rss>