선형대수학 입문/선형 연립방정식 문서 원본 보기
←
선형대수학 입문/선형 연립방정식
둘러보기로 이동
검색으로 이동
문서 편집 권한이 없습니다. 다음 이유를 확인해주세요:
요청한 명령은 다음 권한을 가진 사용자에게 제한됩니다:
사용자
.
문서의 원본을 보거나 복사할 수 있습니다.
== 선형 연립방정식의 행렬꼴 == 우리는 선형 연립방정식을 행렬꼴로 나타내기 전에 그것이 뭔지 정의할 필요가 있습니다. {{인용문| '''정의 2.1. 선형 연립방정식''' 자연수 <math>n</math>개의(우리 논의에서 0은 자연수가 아니다.) 미지수 <math>x_1,x_2,\ldots,x_n</math>의 선형 연립방정식은 다음 형태의 방정식 묶음으로 나타낸다. <math display=block> \begin{cases} a_{11}x_1+a_{12}x_2+\cdots+a_{1n}x_n=b_1,\\ a_{21}x_1+a_{22}x_2+\cdots+a_{2n}x_n=b_2,\\ \quad\qquad\qquad\qquad\vdots\\ a_{m1}x_1+a_{m2}x_2+\cdots+a_{mn}x_n=b_m, \end{cases} </math> 여기서 <math>a_{ij}</math>와 <math>b_k</math>는 어떤 상수다.}} {{인용문| '''참고''' * 몇몇 다른 정의에서는 하나의 선형 방정식(계수가 1인 방정식) 역시도 선형 연립방정식으로 취급하는 경우가 있지만, 하나의 선형 방정식은 풀기 쉽고, 이런 경우는 우리의 관심에서 멀기 때문에, 우리는 이걸 고려하지 않을 것이다. }} 앞으로 선형 연립방정식을 기술할 때 '일관적'(consistent), 그리고 '모순적'(혹은 비일관적, inconsistent)이라는 단어를 사용할 것입니다. [[파일:Three Lines.svg|섬네일|두 미지수가 있고, 세 개의 선형 방정식(선)이 있는 이 그림은 세 선이 동시에 지나는 점이 없으므로 일관적이지 않다.]] {{인용문| '''정의 2.2. (선형 연립방정식의 일관성)''' 선형 연립방정식은 적어도 하나의 해가 있으면 일관적이다. 그렇지 않으면 모순적이다.(따라서 해가 하나도 없으면 모순적이다.)}} {{인용문| '''참고''' * 우리가 나중에 볼 것이지만, 선형 연립방정식은 해가 전혀 없거나, 하나만 있거나, 무수히 많다. 따라서 선형 연립방정식이 일관적이면 해가 유일하거나 무수히 많은 것은 동치다. }} {{인용문| '''예시 (일관적인 선형 연립방정식)''' 다음 선형 연립방정식을 고려하자. <math display=block> \begin{cases} x+2y=3&\quad(1)\\ 3x+6y=9&\quad(2) \end{cases}. </math> 여기서 <math display=block> (2)\iff 3x+6y=9\iff (3/3)x+(6/3)y\iff (9/3)\iff x+2y=3\iff (1) </math> 이므로, 이 선형 연립방정식의 해는 <math>x+2y=3</math>를 만족하는 <math>(x,y)</math>이다. 따라서 이 선형 연립방정식의 해는 무수하게 많고, 그러므로 일관적이다. }} {{인용문| '''예시 (모순적인 선형 연립방정식)''' 다음 선형 연립방정식을 고려하자. <math display=block> \begin{cases} x+2y=3&\quad(1)\\ x+2y=4&\quad(2) \end{cases}. </math> <math>(2)</math>에 <math>(1)</math>을 대입하면, <math display=block> 3=4 </math> 이고, 말이 안되는 소리이므로 두 방정식을 만족하는 <math>(x,y)</math>는 존재하지 않는다. 이는 선형 연립방정식의 해가 없다는 것이고, 따라서 이 선형 연립방정식은 모순적이다. }} {{인용문| '''예시 (선형 연립방정식의 응용)''' 10병의 오렌지 주스가 닭, 오리, 거위에게 배당한다고 하자. 닭과 오리에게 똑같은 개수의 병이 주어지고, 거위는 닭과 오리에 비해 한 병 더 많다고 하면 각각의 동물에게 얼마의 병이 배당되었는가? 답: 닭, 오리, 거위에게 할당된 오렌지 주스 병의 개수를 각각 <math>c,d, g</math>라고 하자. 그러면 주어진 상황과 조건에 따라 우리는 다음과 같은 선형 연립방정식을 세울 수 있다. <math display=block> \begin{cases} c+d+g=10&\quad(1)\\ c=d&\quad(2)\\ g=c+1&\quad(3) \end{cases} </math> 여기서 <math>c,d,g</math>는 모두 음이 아닌 정수다. <math>(2)</math>와 <math>(3)</math>을 <math>(1)</math>에 대입하면 <math display=block> c+c+c+1=10\implies c=3. </math> 따라서 <math>d=c=3</math>이고 <math>g=c+1=4</math>이다. 즉, 닭과 오리는 각각 3개의 오렌지 주스 병이 배당되었고, 거위는 4개의 오렌지 주스 병이 배당되었다. }} {{예제| <quiz display=simple> {Choose SLE(s) from the following. |type="[]"} - <math display=block>\begin{cases}x^2+y^2=1\\2x+3y=4\end{cases}</math> || <math>x^2+y^2=1</math> is nonlinear (i.e. not in the form of the equations in the definition of SLE) + <math display=block>\begin{cases}x=1\\y=2\end{cases}</math> || it may be also expressed as <math display=block>\begin{cases}x+0y=1\\0x+y=2\end{cases}</math>, and then we can see that it is SLE + <math display=block>\begin{cases}x=1\\x=2\end{cases}</math> || it is SLE, even if it is inconsistent + <math display=block>\begin{cases}\pi x+\sqrt \pi y=\ln 2\\x+y\sqrt 2=2\end{cases}</math> || even if some coefficients are irrational, this is still SLE {The final score of each student (the highest possible score is <math>100</math>) is a weighted average of the score of the student in test 1 and test 2 (full marks of test 1 and test 2 are both <math>100</math>). Student A gets <math>90</math> and <math>80</math> marks from test 1 and test 2 respectively, and student B gets <math>75</math> and <math>100</math> marks from test 1 and test 2 respectively. Suppose the weighting on the scores of test 1 and test 2 are <math>w_1</math> and <math>w_2</math> respectively. It is given that the final score of student A is <math>85.23</math> exactly. Which of the following is (are) true? |type="[]"} - the given information is not sufficient to compute <math>w_1</math> and <math>w_2</math> - one possible allocation of weightings is given by <math>(w_1,w_2)=(0.6,0.390375)</math> || this is impossible since <math>w_1+w_2\ne 1</math> - the final score of student B is the same as that of student A + the final score of student B is strictly higher than that of student A || from the situation and the given condition, we have the following SLE: || <math display=block>\begin{cases}90w_1+80w_2=\\w_1+w_2=1\end{cases}</math> || solving this, we get <math>w_1=0.523</math> and <math>w_2=0.477</math> || in particular, we have <math>w_1+w_2=1</math>, since the final score is computed based on test 1 and test 2 only, so the sum of weightings must be <math>1</math> || after obtaining the weighting, we can compute the final score of student B, which is <math>75(0.523)+100(0.477)=86.925</math> - the final score of student B is strictly lower than that of student A - we do not know whether the final score of student B is higher, lower than, or the same as that of student A </quiz> }} 선형 연립방정식을 정의했으므로, 우리는 이것을 다양한 방법의 행렬 꼴로 표현할 수 있다. 그리고 그것을 다음과 같이 정의한다. {{인용문| '''정의 2.3 (계수행렬과 첨가행렬)''' <math display=block> \begin{cases} a_{11}x_1+a_{12}x_2+\cdots+a_{1n}x_n=b_1,\\ a_{21}x_1+a_{22}x_2+\cdots+a_{2n}x_n=b_2,\\ \quad\qquad\qquad\qquad\vdots\\ a_{m1}x_1+a_{m2}x_2+\cdots+a_{mn}x_n=b_m \end{cases} </math> 를 미지수 <math>x_1,x_2,\ldots,x_n</math>가 주어지고, <math>a_{ij}</math>와 <math>b_k</math>는 어떤 상수인 선형 연립방정식을 생각하자. 행렬 <math>{\color{green}(a_{ij})_{m\times n}}</math>은 연립방정식의 계수 행렬이라 하고, 행렬 <math display=block> \left( \begin{array}{cccc|c} a_{11}&a_{12}&\cdots&a_{1n}&b_1\\ a_{21}&a_{22}&\cdots&a_{2n}&b_2\\ \vdots&\vdots&\ddots&\vdots&\vdots\\ a_{m1}&a_{m2}&\cdots&a_{mn}&b_n \end{array} \right) </math> 은 연립방정식의 첨가행렬이라고 한다.}} {{인용문| '''참고''' * 첨가행렬에서 수직선은 선택적인 것으로, 상수를 '<math>=</math>'의 왼쪽과 오른쪽으로 나누기 위해 선형 연립방정식에서 사용한다. * 이 연립방정식은 다음과 똑같다. <math display=block> \underbrace{\begin{pmatrix} a_{11}&a_{12}&\cdots&a_{1n}\\ a_{21}&a_{22}&\cdots&a_{2n}\\ \vdots&\vdots&\ddots&\vdots\\ a_{m1}&a_{m2}&\cdots&a_{mn} \end{pmatrix}}_{\text{계수}\;A\text{표현}} \underbrace{\begin{pmatrix} x_1\\ x_2\\ \vdots\\ x_n \end{pmatrix}}_{\text{미지수}\;\mathbf x\text{표현}} = \underbrace{\begin{pmatrix} b_1\\ b_2\\ \vdots\\ b_n \end{pmatrix}}_{\text{상수}\;\mathbf b\text{표현}}, </math> :이는 <math>A\mathbf x=\mathbf b</math>로 다시 쓸 수 있다. * 미지수가 어떤 상수인지 특정할 수 없기 때문에 미지수만 표현하는 것은 중요하지 않다. 따라서 첨가행렬은 연립방정식을 푸는데 필수적인 정보를 준다.}} {{인용문| '''예시(계수행렬과 첨가행렬)''' 선형 연립방정식 <math display=block> \begin{cases} w+x+y=1\\ w+z=3\\ x+z=8 \end{cases} </math> 을 고려하자. 이것을 달리 표현하면 다음과 같다. <math display=block> \begin{cases} w+x+y+0z=1\\ w+0x+0y+z=3\\ 0w+x+0y+z=8 \end{cases} </math> 선형 연립방정식의 계수행렬은 <math display=block> \begin{pmatrix} 1&1&1&0\\ 1&0&0&1\\ 0&1&0&1 \end{pmatrix} </math> 이고, 선형 연립방정식의 첨가행렬은 <math display=block> \left( \begin{array}{cccc|c} 1&1&1&0&1\\ 1&0&0&1&3\\ 0&1&0&1&8 \end{array} \right) </math> 이다. 우리는 이 선형 연립방정식을 또 <math display=block> \begin{pmatrix} 1&1&1&0\\ 1&0&0&1\\ 0&1&0&1 \end{pmatrix} \begin{pmatrix} w\\ x\\ y\\ z \end{pmatrix} =\begin{pmatrix} 1\\ 3\\ 8 \end{pmatrix} </math> 로 표현할 수 있고 또 <math>A\mathbf x=\mathbf b</math>로 사용할 수 있다. }} {{예제| <quiz display=simple> {Choose SLE(s) from the following. |type="[]"} - <math>(a_{ij})_{3\times 3}</math> - <math>\left(\begin{array}{cc|c}1&2&3\\ 4&5&6\end{array}\right)</math> + <math>\begin{pmatrix}1&2\\ 4&5\end{pmatrix}\begin{pmatrix}x_1\\x_2\end{pmatrix}=\begin{pmatrix}3\\6\end{pmatrix}</math> - <math>A\mathbf x=\mathbf b</math> in which <math>A</math> is a <math>3\times 2</math> matrix, <math>\mathbf x,\mathbf b</math> are <math>3\times 1</math> matrices. || the matrix product <math>A\mathbf x</math> is undefined since the number of columns of <math>A</math> is different from the number of rows of <math>\mathbf x</math> {A SLE is represented by the {{colored em|augmented}} matrix <math display=block>\begin{pmatrix}1&0&0&0&0&0&0\\0&1&0&0&0&0&0\\0&0&1&0&0&0&6\\0&0&0&0&1&0&3\\0&0&0&1&0&0&8\\0&0&0&0&0&1&2\end{pmatrix}.</math> Choose correct statement(s). |type="[]"} + one possible solution is <math>(x_1,x_2,x_3,x_4,x_5,x_6)=(0,0,6,3,8,2)</math> || this is true since the naming of variables is arbitrary + one possible solution is <math>(x_1,x_2,x_3,x_4,x_5,x_6)=(0,0,6,8,3,2)</math> || this is true since the naming of variables is arbitrary - the size of the coefficient matrix of this SLE is <math>6\times 7</math> || the size of the {{colored em|augmented}} matrix is <math>6\times 7</math> instead || the size of the coefficient matrix is <math>6\times 6</math> + the SLE in the question has unique solution || for each naming of variables, the solution is unique || the naming of variables of the SLE in the question is one of all possible naming of variables || so the solution is unique for the SLE in the question {Which of the following is (are) the augmented matrix (matrices) representing the SLE <math display=block>\begin{cases}x+2y=3\\6x+8y+2z=0\end{cases}</math> ? |type="[]"} - the augmented matrix of this SLE does not exist + <math>\begin{pmatrix}1&2&0&3\\ 6&8&2&0\end{pmatrix}</math> - <math>\begin{pmatrix}1&2&3\\ 6&8&2\end{pmatrix}</math> - <math>\begin{pmatrix}1&2&0\\ 6&8&2\end{pmatrix}</math> {Choose augmented matrix (matrices) that represents an inconsistent SLE from the following. |type="[]} + <math>\begin{pmatrix}1&1&1\\ 6&6&0\end{pmatrix}</math> || using <math>x,y</math> to name the variables, || it represents <math>\begin{cases}x+y=1\\6x+6y=0\end{cases}</math> + <math>\begin{pmatrix}1&1\\ 2&3\end{pmatrix}</math> || using <math>x</math> to name the variable, || it represents <math>\begin{cases}x=1\\2x=3\end{cases}</math> + <math>I_2</math> || using <math>x</math> to name the variable, || it represents <math>\begin{cases}x=0\\0=1\end{cases}</math> + <math>I_3</math> || using <math>x,y</math> to name the variables, || it represents <math>\begin{cases}x=0\\y=0\\0=1\end{cases}</math> - <math>O_{2\times 2}</math> || using <math>x,y</math> to name the variables, || it represents <math>\begin{cases}0x+0y=0\\0x+0y=0\end{cases}</math> || the solution is <math>(x,y)</math> in which <math>x,y</math> are real numbers </quiz> }} ==가우스-요르단 소거법== {{인용문| '''정의 2.4. (기본 행 연산)''' 행렬에서 적용할 수 있는 기본 행 연산에는 세 가지가 있다: * 두 행의 치환 * 행에 0이 아닌 스칼라의 곱셈 * 한 행에 다른 행의 스칼라배 더하기 }} {{인용문| '''참고''' 기본 행 연산에서 다음과 같은 표기를 따를 것이다: * <math>\mathbf r_1,\mathbf r_2,\ldots,\mathbf r_m</math>: <math>m\times n</math> 행렬의 행 (행은 원래 벡터이므로 '''볼드체'''를 쓴다.) * <math>\mathbf r_{\color{blue}i} {\color{green}\leftrightarrow} \mathbf r_{\color{red}j}</math>: <math>\color{blue}i</math>번째 행과 <math>\color{red}j</math>번째 행을 치환 * <math>{\color{green}k}\mathbf r_i\to \mathbf r_i</math>: <math>i</math>번째 행에 0이 아닌 스칼라 <math>\color{green}k</math>를 곱하기 * <math>{\color{purple}k}\mathbf r_{\color{red}j} {\color{green}+}\mathbf r_{\color{blue}i}\to \mathbf r_{\color{blue}i}</math>: <math>\color{blue}i</math>번째 행에 <math>\color{purple}k</math>배를 한 <math>\color{red}j</math>번째 행 더하기 }} {{인용문| '''정의 2.5. (행 동치, Row equivalence)''' 같은 크기의 두 행렬에 대하여 한 행렬에다가 어떤 기본 행연산을 실행해서 다른 행렬을 만들어낼 수 있으면 서로 '''행 동치'''다. }} {{인용문| '''참고''' * 기본 행연산은 (성질에 의해)가역적이라서 행렬 <math>B</math>를 기본 행연산으로 행렬 <math>A</math>로 바꿀 수 있으면 <math>A</math>도 기본 행연산을 통해 <math>B</math>로 만들어낼 수 있다.(각각의 연산의 역을 찾고 순서에 맞게 잘 정리해서 얻어낼 수 있다.) :* 따라서 행 동치를 증명하기 위해서 어떤 한 쪽의 행렬이 다른 쪽의 행렬로 기본 행연산을 하면 같아진다는 것만 증명하면 된다.(물론 두 행렬의 크기는 같아야한다.) 이에 대한 증명은 아래에서 할 것이다. }} {{인용문| '''예시 (Demonstration of three types of EROs)''' 행렬 <math display=block> A=\begin{pmatrix} {\color{red}1}&{\color{red}2}&{\color{red}3}\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{green}7}&{\color{green}8}&{\color{green}9} \end{pmatrix} </math> 을 생각하자. 그러면 기본 행연산은 다음을 같이 연산한다: <math display=block> \begin{align} \begin{pmatrix} {\color{red}1}&{\color{red}2}&{\color{red}3}\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{green}7}&{\color{green}8}&{\color{green}9} \end{pmatrix} &\overset{\mathbf r_1\leftrightarrow\mathbf r_3}\to \begin{pmatrix} {\color{green}7}&{\color{green}8}&{\color{green}9}\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{red}1}&{\color{red}2}&{\color{red}3} \end{pmatrix}\\ &\overset{\pi\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} {\color{green}7}\pi&{\color{green}8}\pi&{\color{green}9}\pi\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{red}1}&{\color{red}2}&{\color{red}3} \end{pmatrix}\\ &\overset{2\mathbf r_2+\mathbf r_3\to\mathbf r_3}\to \begin{pmatrix} {\color{green}7}\pi&{\color{green}8}\pi&{\color{green}9}\pi\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ 2({\color{blue}4})+{\color{red}1}&2({\color{blue}5})+{\color{red}2}&2({\color{blue}6})+{\color{red}3}\\ \end{pmatrix}\\ &\overset{-2\mathbf r_2+\mathbf r_3\to\mathbf r_3}\to \begin{pmatrix} {\color{green}7}\pi&{\color{green}8}\pi&{\color{green}9}\pi\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{red}1}&{\color{red}2}&{\color{red}3}\\ \end{pmatrix}\\ &\overset{(1/\pi)\mathbf r_1\rightarrow\mathbf r_1}\to \begin{pmatrix} {\color{green}7}&{\color{green}8}&{\color{green}9}\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{red}1}&{\color{red}2}&{\color{red}3}\\ \end{pmatrix}\\ &\overset{\mathbf r_1\rightarrow\mathbf r_3}\to \begin{pmatrix} {\color{red}1}&{\color{red}2}&{\color{red}3}\\ {\color{blue}4}&{\color{blue}5}&{\color{blue}6}\\ {\color{green}7}&{\color{green}8}&{\color{green}9}\\ \end{pmatrix}=A. \end{align} </math> 여기에 보여진 각각의 행렬은 각자 <math>A</math>로 가는 기본 행연산이 주어져있기 때문에 <math>A</math>와 행 동치를 이루고, <math>A</math>와 같은 크기를 가진다.(그리고 기본 행연산에 대한 이전으로 돌아가는 역산도 보여준다.) }} {{예제| <quiz display=simple> {Which of the following is (are) row equivalent to <math>I_3</math>, the identity matrix with size <math>3\times 3</math>? |type="[]"} - <math>I_4</math> + <math>\begin{pmatrix}0&1&0\\1&0&0\\0&0&1\\\end{pmatrix}</math> || it can be obtained from <math>I_3</math> by performing the ERO <math>\mathbf r_1\leftrightarrow \mathbf r_2</math> || {{colored em|remark}}: it is an example of {{colored em|elementary matrix}} of type I, which will be defined later + <math>\begin{pmatrix}9&0&0\\0&1&0\\0&0&1\\\end{pmatrix}</math> || it can be obtained from <math>I_3</math> by performing the ERO <math>9\mathbf r_1\to \mathbf r_1</math> || {{colored em|remark}}: it is an example of {{colored em|elementary matrix}} of type II, which will be defined later + <math>\begin{pmatrix}1&0&0\\-7&1&12\\-8&0&1\\\end{pmatrix}</math> || it can be obtained from <math>I_3</math> by performing the EROs <math>-7\mathbf r_1+\mathbf r_2\to \mathbf r_2,12\mathbf r_3+\mathbf r_2\to\mathbf r_2,-8\mathbf r_1+\mathbf r_3\to\mathbf r_3</math>, {{colored em|in this order}} + <math>\begin{pmatrix}0&1&0\\1&0&0\\0&0&1\\\end{pmatrix}\begin{pmatrix}9&0&0\\0&1&0\\0&0&1\\\end{pmatrix}</math> || it equals <math>\begin{pmatrix}0&1&0\\9&0&0\\0&0&1\end{pmatrix}</math> || it can be obtained from <math>I_3</math> by performing the EROs <math>\mathbf r_1\leftrightarrow \mathbf r_2, 9\mathbf r_1\to\mathbf r_1</math>, {{colored em|in this order}} {Choose correct statement(s). |type="[]"} - after performing EROs <math>\mathbf r_1\leftrightarrow\mathbf r_2,\mathbf r_2\leftrightarrow\mathbf r_3,\mathbf r_3\leftrightarrow\mathbf r_1</math>, {{colored em|in this order}}, on a matrix with at least three rows, the resultant matrix is the same as the original matrix || counterexample: after performing these EROs in this order on <math>I_3</math>, we get <math>\begin{pmatrix}1&0&0\\0&0&1\\0&1&0\\\end{pmatrix}\ne I_3</math> - after performing the ERO <math>3\mathbf r_1+\mathbf r_2\to\mathbf r_2</math> on <math>I_2</math>, we get <math>\begin{pmatrix}3&0\\0&1\end{pmatrix}</math> || we get this matrix from <math>I_2</math> by performing the ERO <math>3\mathbf r_1\to\mathbf r_1</math> || after performing this ERO on <math>I_2</math>, we get instead <math>\begin{pmatrix}1&0\\3&1\end{pmatrix}</math> - given two arbitrary EROs, performing them on the same matrix in different orders give the same resultant matrix || counterexample: performing <math>\mathbf r_1\leftrightarrow\mathbf r_2,\mathbf r_1\leftrightarrow\mathbf r_3</math> in this order and performing <math>\mathbf r_1\leftrightarrow\mathbf r_3,\mathbf r_1\leftrightarrow\mathbf r_2</math> on <math>I_3</math> gives || <math>\begin{pmatrix}0&0&1\\1&0&0\\0&1&0\\\end{pmatrix}</math> and || <math>\begin{pmatrix}0&1&0\\0&0&1\\1&0&0\\\end{pmatrix}</math> respectively || actually, this is false in general - given two arbitrary EROs, performing them on the same matrix in different orders give the different resultant matrices || counterexample: performing <math>2\mathbf r_1\to\mathbf r_1,\frac{1}{2}\mathbf r_1\to\mathbf r_1</math> in different orders on <math>I_2</math> gives <math>I_2</math> </quiz> }} {{인용문| '''성질 (기본 행연산의 가역성)''' 행렬<math>B</math>에 적절한 기본 행연산으로 행렬 <math>A</math>를 만들 수 있다면, <math>A</math> 역시도 적절한 기본 행연산으로 <math>B</math>로 바꿀 수 있다.(역과정에서 사용한 기본 행연산은 조금 다를 수 있다.) }} {{증명}} 설명: 각자의 기본 행연산에는 역과정이 있고(다시 말해 모든 기본 행연산은 그 역과정과 같이 존재하고, 이것은 행렬에 영향을 미치지 못한다.), 이의 역과정은 다음과 같다: * 치환 기본 행연산 <math>\mathbf r_i\leftrightarrow \mathbf r_j</math>의 역과정 역시 <math>\mathbf r_i\leftrightarrow \mathbf r_j</math>이다. * 스칼라곱 기본 행연산 <math>k\mathbf r_i\to \mathbf r_i</math>의 역과정은 스칼라곱 기본행연산 <math>\frac{1}{k}\mathbf r_i\to\mathbf r_i</math> (<math>k=0</math>일 때의 이 연산 자체가 정의되어 있지 않다. 이것이 <math>k</math>가 반드시 0이 아니어야 하는 이유고, 따라서 역과정이 존재한다.) * 덧셈 기본 행연산 <math>k\mathbf r_j+\mathbf r_i\to\mathbf r_i</math>의 역과정은 덧셈 기본 행연산 <math>-k\mathbf r_j+\mathbf r_i\to\mathbf r_i</math>이다. {{증명 끝}} {{인용문| '''예시 (각각의 기본 행연산의 역과정들)''' * 치환: <math display=block> \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \overset{\mathbf r_1\leftrightarrow\mathbf r_2}\to \begin{pmatrix} {\color{blue}3}&{\color{blue}4}\\ {\color{red}1}&{\color{red}2}\\ \end{pmatrix} \overset{\mathbf r_1\leftrightarrow\mathbf r_2}\to \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} </math> * 스칼라곱: <math display=block> \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \overset{{\color{green}k}\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} {\color{red}1}{\color{green}k}&{\color{red}2}{\color{green}k}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \overset{{\color{green}\frac{1}{k}}\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} {\color{red}1}{\color{green}\cancel k}\cdot{\color{green}\frac{1}{\cancel k}}&{\color{red}2}{\color{green}\cancel k}\cdot{\color{green}\frac{1}{\cancel k}}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} = \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \quad (k\ne 0) </math> * 덧셈: <math display=block> \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \overset{{\color{green}k}\mathbf r_2+\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} {\color{red}1}+{\color{blue}3}{\color{green}k}&{\color{red}2}+{\color{blue}4}{\color{green}k}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} \overset{{\color{green}-k}\mathbf r_2+\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} {\color{red}1}\cancel{+{\color{blue}3}{\color{green}k}-{\color{blue}3}{\color{green}k}} &{\color{red}2}\cancel{+{\color{blue}4}{\color{green}k}-{\color{blue}4}{\color{green}k}}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} = \begin{pmatrix} {\color{red}1}&{\color{red}2}\\ {\color{blue}3}&{\color{blue}4}\\ \end{pmatrix} </math> }} {{예제| <quiz display=simple> {Choose correct ERO(s) that is (are) the reverse process of the ERO <math>2^{-1}\mathbf r_3\to\mathbf r_3</math>. |type="[]"} - <math>2\mathbf r_1\to\mathbf r_1</math> + <math>2\mathbf r_3\to\mathbf r_3</math> - <math>2^{-1}\mathbf r_3\to\mathbf r_3</math> - <math>2^{-1}\mathbf r_1\to\mathbf r_1</math> {Choose correct ERO(s) that is (are) the reverse process of the ERO <math>2^{-1}\mathbf r_3+\mathbf r_1\to\mathbf r_1</math>. |type="[]"} - <math>-2^{-1}\mathbf r_1+\mathbf r_3\to\mathbf r_3</math> - <math>-2\mathbf r_1+\mathbf r_3\to\mathbf r_3</math> + <math>-2^{-1}\mathbf r_3+\mathbf r_1\to\mathbf r_1</math> - <math>-2\mathbf r_3+\mathbf r_1\to\mathbf r_1</math> </quiz> }} {{인용문| '''성질 (행 동치와 해)''' <math>A\mathbf x=\mathbf b</math>와 <math>C\mathbf y=\mathbf d</math>인 방정식 개수와 변수의 개수가 같은 두 개의 선형 연립방정식을 생각해보자. 첨가행렬 <math>(A|\mathbf b)</math>와 <math>(C|\mathbf d)</math>가 행 동치이면, 두 개의 연립방정식은 같은 해들을 갖는다. }} {{인용문| '''증명''' 설명: 하나의 기본 행연산으로 해들이 바뀌지 않음을 보이기만 하면 충분하다. 예를 들어 * 치환: <math display=block> \begin{cases} x+2y=3\\ 4x+5y=6\\ \end{cases} \text{ and } \begin{cases} 4x+5y=6\\ x+2y=3\\ \end{cases} \text{비 교}\; </math> * 스칼라 곱셈: <math display=block> \begin{cases} x+2y=3\\ 4x+5y=6\\ \end{cases} \text{ and } \begin{cases} kx+2ky=3k\\ 4x+5y=6\\ \end{cases} \Leftrightarrow \begin{cases} x+2y=3\\ 4x+5y=6\\ \end{cases} \quad (k\ne 0) \text{비 교}\; </math> * 덧셈: <math display=block> \begin{cases} x+2y=3\\ 4x+5y=6\\ \end{cases} \text{ and } \begin{cases} x+2y=3&\quad(1)\\ (4+k)x+(5+2k)y=6+3k&\quad(2)\\ \end{cases} \text{비 교}\; </math> <math>(1)-k(2):</math> <math display=block> (4+k-k)x+(5+2k-2k)y=6+3k-3k \iff 4x+5y=6 </math> }} {{예제| Consider three row equivalent matrices <math display=block> A=\begin{pmatrix} 7&5&3&4\\ 6&2&3&2\\ 4&8&1&6\\ \end{pmatrix}, B=\begin{pmatrix} 1&0&0&-1\\ 0&1&0&1\\ 0&0&1&2\\ \end{pmatrix}, C=\begin{pmatrix} 7&5&3&4\\ 0&16&-3&10\\ 0&0&1&2 \end{pmatrix}. </math> <quiz display=simple> {Solve the SLE <math display=block>\begin{cases}7x+5y+3z&=4\\6x+2y+3z&=2\\4x+8y-z&=6\end{cases}.</math> |type="{}"} Its unique solution: <math>x=</math>{ -1 _2} <math>y=</math>{ 1 _2} <math>z=</math>{ 2 _2} || the augmented matrix representing this SLE is <math>A</math> || since <math>A</math> and <math>B</math> are row equivalent, || its solution set is the same as the SLE represented by <math>B</math> (as augmented matrices) || then we can directly read the solution set from the SLE {Solve the SLE <math display=block>\begin{cases}3x+5y+7z&=4\\-3x+16y&=10\\x&=2\end{cases}.</math> |type="{}"} Its unique solution: <math>x=</math>{ 2 _2} <math>y=</math>{ 1 _2} <math>z=</math>{ -1 _2} || by some rearranging, the SLE becomes || <math display=block>\begin{cases}7z+5y+3x=4\\+16y-3x&=10\\x&=2\end{cases}.</math> || then we can see that this SLE is represented by the augmented matrix <math>C</math>, with 1st, 2nd, 3rd column representing the coefficients of <math>z,y,x</math> respectively || since <math>C</math> and <math>B</math> are row equivalent, the solution set of this SLE is the same as the SLE represented by the augmented matrix <math>B</math> with 1st, 2nd, 3rd column representing the coefficients of <math>z,y,x</math> respectively </quiz> }} {{인용문| '''정의 2.6. (선행성분)''' 행렬의 행의 선행성분은 0이 아닌 가장 왼쪽에 있는 성분이다. }} {{인용문| '''예시''' 아래 행렬의 첫번째, 두번째 세번째 행의 선행성분은 <math display=block> \begin{pmatrix} 0&{\color{blue}2}&3\\ 0&0&{\color{blue}3}\\ {\color{blue}8}&6&2\\ \end{pmatrix} </math> 각각 <math>2,3,8</math>이다. }} {{예제| <quiz display=simple> {What is the leading entry of the first row of <math>O_{3\times 3}</math>? |type="()"} - 0 - 1 - 2 - 3 + it does not exist {What is the leading entry of the first row of <math>I_{3\times 3}</math>? |type="()"} - 0 + 1 - 2 - 3 - it does not exist </quiz> }} {{인용문| '''정의 2.7. (사다리꼴행렬)''' 다음 조건을 만족하는 행렬은 행사다리꼴행렬(혹은 사다리꼴행렬, 약자 REF)이다. # 모든 성분이 <math>0</math>인 행은 (존재한다면)행렬의 가장 밑자리에 놓여있다. # <math>0</math>이 아닌 성분이 하나라도 포함된 행의 선행성분은 반드시 항상 위 행렬의 선행성분의 오른쪽에 있다. }} {{인용문| '''정의 2.8. (기약행사다리꼴행렬)''' 다음 조건을 만족하는 행렬을 기약행사다리꼴행렬(Reduced-Row Echelon Form matrix, RREF)이라 한다. # 사다리꼴 행렬이다. # <math>0</math>이 아닌 성분이 하나라도 포함된 행의 선행성분이 <math>{\color{green}1}</math>이다.(이를 선행 1(leading one)이라고 부른다.) # 각각의 선행 1에 대해서 같은 열의 다른 모든 성분은 0이다. }} {{인용문| '''예시(REF와 RREF)''' * RREF가 아닌 REF: <math display=block> \begin{pmatrix} {\color{blue}3}&2&1&1\\ 0&{\color{blue}2}&9&4\\ 0&0&0&{\color{blue}7}\\ 0&0&0&0\\ 0&0&0&0\\ \end{pmatrix}, \begin{pmatrix} {\color{blue}3}&2&1&1\\ 0&{\color{blue}2}&9&4\\ 0&0&0&{\color{blue}7}\\ 0&0&0&0\\ 0&0&0&0\\ \end{pmatrix}, \begin{pmatrix} 0&0&{\color{blue}3}&2&1&0\\ 0&0&0&0&0&{\color{blue}2}\\ 0&0&0&0&0&0\\ \end{pmatrix} </math> * RREF 행렬들(그리고 당연하게 REF이기도 하다.): <math display=block> \begin{pmatrix} {\color{blue}1}&{\color{green}0}&7&{\color{green}0}&1\\ {\color{green}0}&{\color{blue}1}&9&{\color{green}0}&2\\ {\color{green}0}&{\color{green}0}&0&{\color{blue}1}&9\\ {\color{green}0}&{\color{green}0}&0&{\color{green}0}&0\\ \end{pmatrix}, \begin{pmatrix} {\color{blue}1}&{\color{green}0}&7&{\color{green}0}&1\\ {\color{green}0}&{\color{blue}1}&9&{\color{green}0}&2\\ {\color{green}0}&{\color{green}0}&0&{\color{blue}1}&9\\ \end{pmatrix}, \begin{pmatrix} {\color{blue}1}&{\color{green}0}&7&{\color{green}0}\\ {\color{green}0}&{\color{blue}1}&9&{\color{green}0}\\ {\color{green}0}&{\color{green}0}&0&{\color{blue}1}\\ \end{pmatrix}, \begin{pmatrix} 0&{\color{blue}1}&{\color{green}0}&{\color{green}0}\\ 0&{\color{green}0}&{\color{blue}1}&{\color{green}0}\\ 0&{\color{green}0}&{\color{green}0}&{\color{blue}1}\\ \end{pmatrix}, I, O </math> * REF가 아닌 행렬들(마찬가지로 RREF도 아니다.): <math display=block> \begin{pmatrix} 1&2&1&1\\ {\color{red}0}&{\color{red}0}&{\color{red}0}&{\color{red}0}\\ 0&1&9&4\\ 0&0&0&1\\ 0&0&0&0\\ \end{pmatrix}, \begin{pmatrix} 0&0&0&0&0&{\color{red}1}\\ 0&0&{\color{red}1}&2&1&0\\ 0&0&0&0&0&0\\ \end{pmatrix}, \begin{pmatrix} {\color{red}1}&2&3\\ {\color{red}4}&5&6\\ {\color{red}7}&8&9\\ \end{pmatrix}, \begin{pmatrix} 0&0&{\color{red}1}\\ 0&{\color{red}1}&0\\ {\color{red}1}&0&0\\ \end{pmatrix}, \begin{pmatrix} {\color{red}1}&0&0\\ 0&{\color{red}1}&0\\ 0&{\color{red}1}&0\\ \end{pmatrix} </math> }} {{예제| <quiz display=simple> {Choose RREF(s) from the following. |type="[]"} - <math>\begin{pmatrix}1&0\\1&0\\0&0\end{pmatrix}</math> - <math>\begin{pmatrix}1&0\\1&0\\0&1\end{pmatrix}</math> - <math>\begin{pmatrix}2&0&2&2\\0&2&0&0\\0&0&0&0\end{pmatrix}</math> + <math>\begin{pmatrix}2&0\\0&2\\0&0\end{pmatrix}</math> - <math>\begin{pmatrix}1&2&0\\0&0&1\\0&2&0\end{pmatrix}</math> {How many possible values are there for <math>x</math> such that the matrix <math display=block>\begin{pmatrix}x&0&0&1\\0&x&2x&0\\x&x&1&3\end{pmatrix}</math> is in RREF? |type="()"} + 0 || if <math>x\ne 0</math>, the colored <math>x</math> in <math>\begin{pmatrix}{\color{blue}x}&0&0&1\\0&{\color{blue}x}&2x&0\\{\color{blue}x}&x&1&3\end{pmatrix}</math> are leading entries || then it is not RREF, since the leading entry of 3rd row is on the left of the leading entry of 2nd row || if <math>x=0</math>, the 2nd row is zero row, which does not lie at the bottom of the matrix, so it is also not RREF in this case || it follows that this matrix is impossible to be RREF - 1 - 2 - 3 - infinitely many {Consider the matrix <math display=block>\begin{pmatrix}y&0&0&0&0&1&0&0\\0&0&0&0&0&0&1&0\\0&0&0&0&0&0&0&0\\0&0&0&0&y&0&y&0\\0&0&0&y&0&0&0&y\\0&0&0&0&0&0&y&y\\\end{pmatrix}</math> How many possible values are there for <math>y</math> such that the matrix is in REF? |type="()"} + 0 || since the zero row is located at the 3rd row, and does not lie at its bottom, the matrix is not REF regardless of the value of <math>y</math> - 1 || consider 3rd row of the matrix - 2 || consider 3rd row of the matrix - 3 || consider 3rd row of the matrix - infinitely many || consider 3rd row of the matrix {Consider the matrix <math display=block>\begin{pmatrix}0&x&0&0\\0&0&y&0\\0&0&0&z\end{pmatrix}</math> How many possible values are there for <math>(x,y,z)</math> such that the matrix is in RREF? |type="()"} - 6 - 7 + 8 || the possible values of <math>(x,y,z)</math> are || <math>(x,y,z)=(0,0,0),(1,0,0),(0,1,0),(0,0,1),(1,1,0),(1,0,1),(0,1,1),(1,1,1)</math> - 9 - infinitely many </quiz> }} {{인용문| '''정의 2.9. (가우스-요르단 소거법)''' 가우스-요르단 소거법은 행렬에 기본 행연산을 적용하여 RREF(기약행사다리꼴행렬) 행렬로 바꾸는 방법이다. 가우스-요르단 소거법은 다음의 과정을 따른다: # 가장 왼쪽에 있는 <math>0</math>이 아닌 성분이 하나라도 포함된 열에 대해서 <math>i</math>열이라고 하자. (필요하다면) <math>i</math>열의 첫번째 성분 <math>a</math>가 <math>0</math>이 아니게 되도록 행을 바꾼다. # 첫번째 행에 <math>a^{-1}</math>을 곱해 <math>i</math>열의 첫번째 성분이 <math>1</math>이 되게 한다. # <math>i</math>열의 <math>0</math>이 아닌 성분 <math>b</math>가 있는 각각의 행에 첫번째 행에 <math>-b</math>를 곱한 것을 더하여 그 성분을 <math>0</math>으로 만든다. # 첫번째 행을 제외한 모든 행의 모든 성분이 <math>0</math>이면 과정은 끝났다. 아니라면, 첫번째 행을 제외하고, 다른 행 중에서 선행성분이 가장 왼쪽에 있는 행을 (필요하다면)두번째 행과 바꾼다. 이 때의 열을 <math>j</math>열이라고 하고, 선행성분을 <math>c</math>라고 하자. # 두번째 행에 <math>c^{-1}</math>를 곱해서 <math>j</math>열의 두번째 행을 <math>1</math>로 만든다. # <math>j</math>열의 각각의 <math>0</math>이 아닌 성분 <math>d</math>가 있는 행에 add 두번째 행에 <math>-d</math>를 곱한 것을 더해서그 성분을 <math>0</math>으로 만든다. # 위 과정을 반복하여 모든 열과 행에 대해 적용하거나, 나머지 행을 0으로 만들어버린다. 그러면 이 결과로 나온 행렬은 RREF 행렬이다. }} {{인용문| '''참고''' * 행렬 <math>A</math>의 RREF 행렬에 대해 기본 행연산을 잘 수행하면 <math>A</math>를 얻을 수 있다. :* 모든 행렬에 가우스-요르단 소거법을 사용할 수 있으므로, 모든 행렬은 RREF 행렬이 존재한다. * 행렬의 RREF 행렬은 유일하다. (증명은 복잡하니 생략한다.) * 가우스-요르단 소거법의 몇몇 다른 정의에서는 과정이 다를 수 있지만, 그래도 여전히 행렬을 그것의 RREF로 바꿀 수 있다. * 인터넷을 뒤져보면 쉽게 가우스-요르단 소거법이나 여타 기본 행연산에 관련된 계산을 하는 [https://www.math.odu.edu/~bogacki/cgi-bin/lat.cgi 이런 사이트]를 찾아볼 수 있을 것이다. }} {{인용문| '''예시 (가우스-요르단 소거법)''' <math display=block> \begin{align} \begin{pmatrix} 0&0&1\\ 3&3&2\\ 8&2&6\\ \end{pmatrix} &\overset{\mathbf r_1\leftrightarrow\mathbf r_2}\to \begin{pmatrix} 3&3&2\\ 0&0&1\\ 8&2&6\\ \end{pmatrix}&\text{과 정 1}\\ &\overset{\frac{1}{3}\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} 1&1&2/3\\ 0&0&1\\ 8&2&6\\ \end{pmatrix}&\text{과 정 2}\\ &\overset{-8\mathbf r_1+\mathbf r_3\to\mathbf r_3}\to \begin{pmatrix} 1&1&2/3\\ 0&0&1\\ 0&-6&2/3\\ \end{pmatrix}&\text{과 정 3}\\ &\overset{\mathbf r_2\leftrightarrow\mathbf r_3}\to \begin{pmatrix} 1&1&2/3\\ 0&-6&2/3\\ 0&0&1\\ \end{pmatrix}&\text{과 정 4}\\ &\overset{-\frac{1}{6}\mathbf r_2\to\mathbf r_2}\to \begin{pmatrix} 1&1&2/3\\ 0&1&-1/9\\ 0&0&1\\ \end{pmatrix}&\text{과 정 5}\\ &\overset{-\mathbf r_2+\mathbf r_1\to\mathbf r_1}\to \begin{pmatrix} 1&0&7/9\\ 0&1&-1/9\\ 0&0&1\\ \end{pmatrix}&\text{과 정 6}\\ &\overset{\frac{1}{9}\mathbf r_3+\mathbf r_2\to\mathbf r_2}{\overset{-\frac{7}{9}\mathbf r_3+\mathbf r_1\to\mathbf r_1}\to} \begin{pmatrix} 1&0&0\\ 0&1&0\\ 0&0&1\\ \end{pmatrix}&\text{과 정 7} \end{align} </math> }} {{예제| <quiz display=simple> {Solve the SLE <math display=block>\begin{cases}x+2y=3\\4x+5y=6\end{cases}</math> using Gauss-Jordan algorithm. |type="{}"} Its unique solution: <math>x=</math> { -1 _2 } <math>y=</math> { 2 _2 } || the augmented matrix representing this SLE is || <math>\begin{pmatrix}1&2&3\\4&5&6\end{pmatrix}\overset{-4\mathbf r_1\to\mathbf r_2}\to\begin{pmatrix}1&2&3\\0&-3&-6\end{pmatrix}\overset{-\frac{1}{3}\mathbf r_2\to\mathbf r_2}\to\begin{pmatrix}1&2&3\\0&1&2\end{pmatrix}\overset{-2\mathbf r_2+\mathbf r_1\to\mathbf r_1}\to\begin{pmatrix}1&0&-1\\0&1&2\end{pmatrix}</math> || then we can read the solution directly from the RREF of the augmented matrix </quiz> Prove that the RREF of the matrix <math display=block> \begin{pmatrix} 1&4&2&1&0&0\\ 3&1&5&0&1&0\\ 0&1&0&0&0&1 \end{pmatrix} </math> is <math display=block> \begin{pmatrix} 1&0&0&-5&2&18\\ 0&1&0&0&0&1\\ 0&0&1&3&-1&-11\\ \end{pmatrix}, </math> and <math display=block> \begin{pmatrix} 1&4&2\\ 3&1&5\\ 0&1&0\\ \end{pmatrix} \begin{pmatrix} -5&2&18\\ 0&0&1\\ 3&-1&-11\\ \end{pmatrix} = \begin{pmatrix} 1&0&0\\ 0&1&0\\ 0&0&1\\ \end{pmatrix} =I_3 </math> }} {{인용문| '''성질 (선형 연립방정식의 해의 개수 결정)''' <math>A\mathbf x=\mathbf b</math>를 <math>m</math>개의 선형 방정식과 <math>n</math>개의 미지수로 구성된 선형 연립방정식이라 하자. 그리고 <math>R</math>을 크기가 <math>m\times (n+1)</math>인 첨가행렬 <math>(A|\mathbf b)</math>의 RREF라고 하자. 그러면 다음을 따른다. * <math>R</math>이 <math>{\color{green}(n+1)}</math>열에서 선행성분(<math>=1</math>)을 가지고 있으면, 이 연립방정식은 모순적이다. * <math>R</math>이 <math>{\color{green}(n+1)}</math>을 제외한 각각의 열에서 선행성분을 가지고 있으면, 이 연립방정식의 해는 유일하다. * <math>R</math>이 <math>{\color{green}(n+1)}</math>열에서 선행성분을 가지고 있지 않고, 이 열을 제외한 각각의 열에 선행성분을 모두 가지고 있는 것이 아니라면, 이 연립방정식의 해는 무수히 많다. }} {{인용문| '''참고''' <math>R</math>은 위 세가지 중 반드시 하나만을 만족해야 하므로, 선형 연립방정식의 해는 반드시 0개이거나(존재하지 않거나), 1개 이거나(유일하거나) 무수히 많다. }} {{인용문| '''예시''' <math>3\times 4</math>의 RREF 첨가행렬로 표현된 선형 연립방정식 <math display=block> \begin{pmatrix} 1&2&0&0\\ 0&0&1&0\\ {\color{green}0}&{\color{green}0}&{\color{green}0}&{\color{green}1}\\ \end{pmatrix} </math> 는 4열에 선행성분을 가지고 있으므로 모순적이다. <math>4\times 4</math>의 RREF 첨가행렬로 표현된 선형 연립방정식 <math display=block> \begin{pmatrix} {\color{green}1}&0&0&2\\ 0&{\color{green}1}&0&3\\ 0&0&{\color{green}1}&3\\ 0&0&0&0\\ \end{pmatrix} </math> 는 앞의 3개의 열은 선행성분을 갖지만, 4열은 그렇지 않으므로 유일한 해를 갖는다. <math>3\times 4</math>의 RREF 첨가행렬로 표현된 선형 연립방정식 <math display=block> \begin{pmatrix} {\color{green}1}&3&0&2\\ 0&0&{\color{green}1}&3\\ 0&0&0&0\\ \end{pmatrix} </math> 는 앞의 3개의 열 중 선행성분이 존재하지 않은 열(2열)이 있고, 동시에 4열에도 선행성분이 존재하지 않으므로 무수히 많은 해를 가진다. 이 행렬을 선형 연립방정식으로 표현하면 <math display=block> \begin{cases} x+3y=2\\ z=3\\ \end{cases} </math> 이다. <math>y</math>를 독립 미지수 <math>t</math>로 두어 <math>y=t</math>로 표현하면 <math>x=2-3t</math>이고 <math>z=3</math>이므로, <math>x</math> 역시도 하나의 해로 결정할 수 없다. }} {{인용문| '''참고''' * 독립 미지수(혹은 자유변수)들은 선행성분이 없는 열에 대응되는 미지수들이다. * 종속 미지수(혹은 기본변수)들은 선행성분이 있는 열에 대응되는 미지수들이다. }} {{예제| It is given that the RREF of the matrix <math display=block> \begin{pmatrix} 1&4&5&6&8\\ 4&6&7&4&4\\ 4&2&3&6&7\\ 1&7&5&3&6\\ \end{pmatrix} </math> is <math display=block> \begin{pmatrix} 1&0&0&0&\frac{-31}{51}\\ 0&1&0&0&\frac{4}{3}\\ 0&0&1&0&\frac{-89}{51}\\ 0&0&0&1&2\\ \end{pmatrix}. </math> Denote the SLE's <math display=block> \begin{cases} 5w+4x+y+6z&=8\\ 7w+6x+4y+4z&=4\\ 3w+2x+4y+6z&=7\\ 5w+7x+y+3z&=6\\ \end{cases}, \begin{cases} 5w+4x+y+6z&=8\\ 7w+6x+4y+4z&=4\\ 3w+2x+4y+6z&=7\\ 5w+7x+y+3z&=6\\ 0w+0x+0y+0z&=0\\ \end{cases}, \begin{cases} 5w+4x+y+6z&=8\\ 7w+6x+4y+4z&=4\\ 3w+2x+4y+6z&=7\\ 5w+7x+y+3z&=6\\ w+x+y+z&=1\\ \end{cases}, </math> by <math>A</math>, <math>B</math> and <math>C</math> respectively. <quiz display=simple> {Does <math>A</math> have no solutions, a unique solution, or infinitely many solutions? |type="()"} - no solutions + a unique solution - infinitely many solutions {Does <math>B</math> have no solutions, a unique solution, or infinitely many solutions? |type="()"} - no solutions + a unique solution - infinitely many solutions {Does <math>C</math> have no solutions, a unique solution, or infinitely many solutions? |type="()"} + no solutions || Putting the unique solution satisfying the first four linear equations <math>(y,x,w,z)(\frac{-31}{51},\frac{4}{3},\frac{-89}{51},2)</math> into the last linear equation, || we get <math>\frac{50}{51}=1</math>, which is always false || so there does not exist a solution such that all five linear equations are satisfied || it follows that <math>C</math> has no solutions - a unique solution - infinitely many solutions </quiz> }} {{인용문| '''정의 2.10. (homogeneous한 선형 연립방정식)''' 선형 연립방정식의 꼴이 <math>A\mathbf x=\mathbf 0</math>이면 homogeneous하다. Homogeneous한 연립방정식은 모든 방정식의 꼴이 미지수항<math>=0</math>이라 해를 가지기 때문에, 자명한 해의 정의에 의해 항상 일관적일 수 밖에 없다. 만약에 다른해가 존재한다면 그 해를 자명하지 않은 해라고 부른다. }} <ref group="해">Homogeneous라는 단어의 마땅한 번역어는 없습니다. Homogeneous라는 단어는 라틴어 homos(같은)+genos(종)이라는 단어의 조합으로 동질의, 동등한, 동차(次)의, 균질한 등으로 번역을 할 수는 있지만, 한 단어로 대응되는 것은 아닙니다.</ref> {{인용문| '''참고''' Homogeneous한 선형 연립방정식은 선형 연립방정식의 해의 개수의 결정에 대한 성질과 homogeneous한 선형 연립방정식들은 일관적이어야만 한다는 사실에 기반하여 homogeneous한 선형 연립방정식들은 해를 갖지 않는다는 가능성 자체를 배제할 수 있기에 단 하나의 유일한 해를 갖거나, 무수히 많은 해들을 가진다. }} {{인용문| '''예시(homogeneous한 선형 연립방정식)''' 선형 연립방정식 <math display=block> \begin{cases} x+y+z&=0\\ 2x+8y+3z&=0\\ 2x+4y+6z&=0\\ \end{cases} </math> 은 homogeneous하고, 따라서 일관적이다. 게다가, 이 선형 연립방정식의 RREF 첨가행렬은 <math display=block> \begin{pmatrix} 1&0&0&0\\ 0&1&0&0\\ 0&0&1&0\\ \end{pmatrix} </math> 이고, 우리는 이 선형 연립방정식의 유일한 해인 <math>(x,y,z)=(0,0,0)</math>를 볼 수 있고, 이것은 자명한 해다. 선형 연립방정식 <math display=block> \begin{cases} x+4y+5z&=0\\ 2x+3y+5z&=0\\ x+2y+3z&=0\\ \end{cases} </math> 은 homogeneous하고, 따라서 일관적이다. 덧붙여, 이 선형 연립방정식의 RREF 첨가행렬은 <math display=block> \begin{pmatrix} 1&0&1&0\\ 0&1&1&0\\ 0&0&0&0\\ \end{pmatrix} </math> 이고, 따라서 이 선형 연립방정식은 3열과 4열의 선행성분이 빠졌기 때문에 수 많은 해를 가진다. 우리는 <math>z</math>를 아무렇게나 둘 수 있다. }} {{인용문| '''성질 (homogeneous한 선형 연립방정식이 비자명한 해를 가지는 충분조건)''' <math>m</math>개의 선형 방정식과 <math>n</math>개의 미지수를 가지고 있는 homogeneous한 연립 방정식은 <math>{\color{green}m<n}</math>이면 반드시 비자명한 해를 가진다. }} {{인용문| 증명 <math>m</math>개의 선형 방정식과 <math>n</math>개의 미지수를 가지고 있는 homogeneous한 연립 방정식의 첨가행렬을 <math>A\mathbf x=\mathbf 0</math>, <math>(A|\mathbf 0)</math>이라 하고, 따라서 이 행렬의 RREF는 <math>(R|\mathbf 0)</math>이라 하자.(여기서 <math>R</math>의 크기는 <math>m</math>개의 선형 방정식과 <math>n</math>개의 미지수를 가지고 있으므로 <math>m\times n</math>이다.) <math>R</math>이 각각의 <math>n</math>개의 열에 대해서 선행성분을 가지고 있으면 <math>R</math>은 적어도 <math>n</math>행을 가지고 있을 것이다. 그러나 <math>R</math>은 <math>m<n</math>인 <math>m</math>개의 행만을 가지므로 모순이다. 따라서, homogeneous한 선형 연립방정식은 유일한 해를 가지지 않는다. 선형 연립방정식은 해가 없거나(homogeneous한 경우 불가능) 유일한 해가 있거나(이 경우에는 불가능) 무수히 많은 해가 있으므로, 이 homogeneous한 선형 연립방정식은 반드시 무수한 해를 가져야 한다. 즉, 비자명한 해를 가진다. }} {{인용문| '''참고''' * 이 성질은 homogeneous한 선형 연립방정식이 비자명한 해를 가진다면 선형 방정식 수가 미지수의 개수보다 딸린다는 주장을 하는 것이 아니다.(성질의 역(逆)) :* 위의 예시(homogeneous한 선형 연립방정식)의 두번째에서 미지수의 개수와 선형 방정식의 개수가 같지만 선형 연립방정식은 여전히 비자명한 해를 갖는다. }} {{인용문| '''예시''' homogeneous한 선형 연립방정식 <math display=block> \begin{cases} x+y+z=0\\ 2x+5y+z=0 \end{cases} </math> 은 반드시 비자명한 해를 가진다. 더불어, 이 선형 연립방정식의 첨가행렬의 RREF는 <math display=block> \begin{pmatrix} 1&0&\frac{4}{3}&0\\ 0&1&-\frac{1}{3}&0 \end{pmatrix} </math> 이다. 이제 <math>z</math>를 독립 미지수 <math>z=t</math>로 두면, 우리는 <math>x=-\frac{4}{3}t,y=\frac{1}{3}t</math>를 얻는다. }} == 해주 == <references group="해" /> [[분류:선형대수학 입문|선형 연립방정식]]
이 문서에서 사용한 틀:
틀:예제
(
원본 보기
)
틀:인용문
(
원본 보기
)
틀:증명
(
원본 보기
)
틀:증명 끝
(
원본 보기
)
선형대수학 입문/선형 연립방정식
문서로 돌아갑니다.
둘러보기 메뉴
개인 도구
로그인
이름공간
문서
토론
한국어
보기
읽기
원본 보기
역사 보기
더 보기
검색
둘러보기
대문
최근 바뀜
임의의 문서로
미디어위키 도움말
특수 문서 목록
도구
여기를 가리키는 문서
가리키는 글의 최근 바뀜
문서 정보