行列と線形写像

○行列の定義

・数や文字を長方形の形に並べて,括弧で囲んだものを行列という.
・横の並びを行といい,上から順に第\( \displaystyle 1 \)行,第\( \displaystyle 2 \)行,\( \displaystyle \cdots \)という.また,縦の並びを列といい,左から順に第\( \displaystyle 1 \)列,第\( \displaystyle 2 \)列,\( \displaystyle \cdots \) という.第\( \displaystyle i \)列を縦ベクトル \( \displaystyle \overset{\xrightarrow{\quad}}{a_i} \) で表すことにすると,行列\( \displaystyle A \)は\( \displaystyle (\overset{\xrightarrow{\quad}}{a_1}, \overset{\xrightarrow{\quad}}{a_2}, \cdots) \)と表される.
・行列の第\( \displaystyle i \)行,第\( \displaystyle j \)列にある成分を\( \displaystyle (i, j) \)成分といい,これを \( \displaystyle a_{ij} \) と表す.行列全体を \( \displaystyle A=(a_{ij}) \)のように表すことで,行列のすべての成分を一括して表現できる.
・\( \displaystyle m \)個の行と\( \displaystyle n \)個の列からなる行列を「\( \displaystyle m \)行\( \displaystyle n \)列の行列」または「\( \displaystyle m \times n \)行列」という.とくに,\( \displaystyle n \)行\( \displaystyle n \)列の行列は\( \displaystyle n \)次の正方行列という.

○行列の和と差,定数(スカラー)倍,積

・ 行列の和と差は同じ型の行列 \( \displaystyle A = (a_{ij}) \) と \( \displaystyle B = (b_{ij}) \) 同士で定義され,対応する成分同士を加減する.
\( \displaystyle \ \ \ \ \ A+B=(a_{ij}+b_{ij}), A-B=(a_{ij}-b_{ij}) \)
・ 行列 \( \displaystyle A = (a_{ij}) \) の定数(スカラー)倍は,行列のすべての成分を定数倍する.
\( \displaystyle \ \ \ \ \ kA = (k \cdot a_{ij}) \ \ \ (k\text{は定数}) \)
・ 行列\( \displaystyle A \)の列数と行列\( \displaystyle B \)の行数が等しいとき,積\( \displaystyle AB \)が定義される.\( \displaystyle A = (a_{ij}) \) が\( \displaystyle m \)行\( \displaystyle n \)列の行列,\( \displaystyle B = (b_{jk}) \)が\( \displaystyle n \)行\( \displaystyle p \)列の行列のとき,積の \( \displaystyle (i, k) \) 成分\( \displaystyle c_{ik} \) は,次のように定義する.
\( \displaystyle \ \ \ \ \ c_{ik}=\sum_{j=1}^{n} a_{ij} \cdot b_{jk} \) (「\( \displaystyle A \)の\( \displaystyle i \)行目」と「\( \displaystyle B \)の\( \displaystyle k \)列目」の対応する成分の積を足し合わせる)

\( \displaystyle (例) 2 \)行\( \displaystyle 2 \)列の行列 \( \displaystyle A = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}, B = \begin{pmatrix} 5 & 6 \\ 7 & 8 \end{pmatrix} \) について,
\( \displaystyle \ \ \ \ \ A+B=\begin{pmatrix} 1+5 & 2+6 \\ 3+7 & 4+8 \end{pmatrix}=\begin{pmatrix} 6 & 8 \\ 10 & 12 \end{pmatrix}=2\begin{pmatrix} 3 & 4 \\ 5 & 6 \end{pmatrix}, \)
\( \displaystyle \ \ \ \ \ A-B=\begin{pmatrix} 1-5 & 2-6 \\ 3-7 & 4-8 \end{pmatrix}=\begin{pmatrix} -4 & -4 \\ -4 & -4 \end{pmatrix}=-4\begin{pmatrix} 1 & 1 \\ 1 & 1 \end{pmatrix}, \)
\( \displaystyle \ \ \ \ \ AB =\begin{pmatrix} \textcolor{red}{1} & \textcolor{red}{2} \\ 3 & 4 \end{pmatrix}\begin{pmatrix} \textcolor{blue}{5} & 6 \\ \textcolor{blue}{7} & 8 \end{pmatrix}=\begin{pmatrix} \textcolor{red}{1} \cdot \textcolor{blue}{5}+\textcolor{red}{2} \cdot \textcolor{blue}{7} & \textcolor{red}{1} \cdot 6+\textcolor{red}{2} \cdot 8 \\ 3 \cdot \textcolor{blue}{5}+4 \cdot \textcolor{blue}{7} & 3 \cdot 6+4 \cdot 8 \end{pmatrix}=\begin{pmatrix} 19 & 22 \\ 43 & 50 \end{pmatrix}, \)
\( \displaystyle \ \ \ \ \ BA =\begin{pmatrix} 5 & 6 \\ 7 & 8 \end{pmatrix}\begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}=\begin{pmatrix} 5 \cdot 1+6 \cdot 3 & 5 \cdot 2+6 \cdot 4 \\ 7 \cdot 1+8 \cdot 3 & 7 \cdot 2+8 \cdot 4 \end{pmatrix}=\begin{pmatrix} 23 & 34 \\ 31 & 46 \end{pmatrix} \)

※正方行列の積について,「任意の\( \displaystyle A, B \)で\( \displaystyle AB=BA \)」は成立しない(交換法則が成り立たない).
※\( \displaystyle \overset{\xrightarrow{\quad}}{b_1}=\begin{pmatrix} 5 \\ 7 \end{pmatrix}, \overset{\xrightarrow{\quad}}{b_2}=\begin{pmatrix} 6 \\ 8 \end{pmatrix} \)とすると,\( \displaystyle AB \)は
\( \displaystyle \ \ \ \ \ AB=A\begin{pmatrix} \overset{\xrightarrow{\quad}}{b_1} & \overset{\xrightarrow{\quad}}{b_2} \end{pmatrix}=\begin{pmatrix} A\overset{\xrightarrow{\quad}}{b_1} & A\overset{\xrightarrow{\quad}}{b_2} \end{pmatrix}=\begin{pmatrix} \begin{pmatrix} 19 \\ 43 \end{pmatrix} & \begin{pmatrix} 22 \\ 50 \end{pmatrix} \end{pmatrix}=\begin{pmatrix} 19 & 22 \\ 43 & 50 \end{pmatrix} \)
のように行列とベクトルの積を\( \displaystyle 2 \)つ並べたものとして考えることもできる.

○単位行列,零行列

・正方行列のうち,対角成分\( \displaystyle a_{ii} \)がすべて\( \displaystyle 1 \)で,それ以外の成分がすべて\( \displaystyle 0 \)であるものを単位行列といい,\( \displaystyle E \)で表す.任意の正方行列\( \displaystyle A \)に対して,それと同じ次数の単位行列\( \displaystyle E \)に対して\( \displaystyle AE=EA=A \)が成り立つ.
・すべての成分が\( \displaystyle 0 \)である行列を零行列といい,\( \displaystyle O \)で表す.任意の正方行列\( \displaystyle A \)に対して,それと同じ次数の正方行列である零行列に対して,\( \displaystyle AO=OA=O \)が成り立つ.

○離散グラフの行列表示

・頂点と辺からなる離散グラフの関係は,隣接行列を用いて表すことができる.
・頂点\( \displaystyle i \)から頂点\( \displaystyle j \)へ辺がある場合に\( \displaystyle (i, j) \)成分を\( \displaystyle 1 \),ない場合に\( \displaystyle 0 \)とする.

(例) \( \displaystyle 3 \)つの頂点\( \displaystyle 1, 2, 3 \)があり,頂点\( \displaystyle 1 \)と\( \displaystyle 2 \),頂点\( \displaystyle 2 \)と\( \displaystyle 3 \)の間に辺があるグラフの隣接行列は次の通り.

○線形写像の定義

幾何ベクトルを幾何ベクトルに移す写像\( \displaystyle f \)で,
\( \displaystyle \vec{x}, \vec{y} \)と定数(スカラー) \( \displaystyle k \) に対して,
(1) \( \displaystyle f(\vec{x}+\vec{y})=f(\vec{x})+f(\vec{y}) \) (2) \( \displaystyle f(k\vec{x})=kf(\vec{x}) \)
を満たすものを線形写像(一次写像)という.

※数の集合を\( \displaystyle 1 \)次元幾何ベクトルと同一視して,「幾何ベクトルから“数”への線形写像」も考えることができる.
※平面から平面,空間から空間など,定義域と値域が同じ集合である写像は,変換ともいう.
※本格的な定義では,定義域と値域は一般的な集合(線形空間)を考え,扱う対象を幾何ベクトルに限定しない.

○線形写像の行列による表現

線形写像は,ある行列\( \displaystyle M \)を用いて \( \displaystyle f(\vec{x})= M\vec{x} \) と表すことができる.

(例)平面の線形変換の例

※例\( \displaystyle 1 \)は定義から直接\( \displaystyle f\left(\begin{pmatrix} x \\ y \end{pmatrix}\right)=\begin{pmatrix} ax \\ by \end{pmatrix}=\begin{pmatrix} a & 0 \\ 0 & b \end{pmatrix}\begin{pmatrix} x \\ y \end{pmatrix} \)と行列表示できることからも,線形写像であることがわかる.

(例,長野県)行列\( \displaystyle \begin{pmatrix} 2 & 1 \\ 1 & 2 \end{pmatrix} \)の表す線形変換(一次変換)によって,直線\( \displaystyle l \)上の点が,すべてこの直線\( \displaystyle l \)上に変換されるとき,このような直線\( \displaystyle l \)の方程式をすべて求める.

○合成変換

・複数の変換を続けて行う変換を,合成変換という.変換\( \displaystyle f \)のあとに続けて変換\( \displaystyle g \)を行う合成変換を\( \displaystyle g \circ f \)で表す.
・ベクトル \( \displaystyle \vec{x} \) にまず行列\( \displaystyle A \)で表される変換を施し,次に行列\( \displaystyle B \)で表される変換を施すとき,\( \displaystyle B(A\vec{x}) = (BA)\vec{x} \) となる.合成変換を表す行列は \( \displaystyle BA \) である.
※「任意の変換\( \displaystyle f, g \)で\( \displaystyle g \circ f = f \circ g \)」は成立しない(交換法則が成り立たない).

(例,名古屋市) \( \displaystyle xy \)平面上で,\( \displaystyle x \)軸に関する対称移動を\( \displaystyle f \),原点\( \displaystyle O \)のまわりの\( \displaystyle 60^\circ \)の回転移動を\( \displaystyle g \)とするとき,合成変換\( \displaystyle g \circ f \)を表す行列を求める.

この記事は役に立ちましたか?

間違い/不具合かな?
と思ったらこちらへ