logo

行空間、列空間、零空間 📂行列代数

行空間、列空間、零空間

定義1

$$ A = \begin{bmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn} \end{bmatrix} $$

行列 $A$に対して、$A$の行から作られる$m$個の$\mathbb{R}^{n}$ベクターは

$$ \begin{align*} \mathbf{r}_{1} =& \begin{bmatrix} a_{11} & a_{12} & \cdots & a_{1n} \end{bmatrix} \\ \mathbf{r}_{2} =& \begin{bmatrix} a_{21} & a_{22} & \cdots & a_{2n} \end{bmatrix} \\ &\vdots \\ \mathbf{r}_{m} =& \begin{bmatrix} a_{m1} & a_{m2} & \cdots & a_{mn} \end{bmatrix} \end{align*} $$

$A$の行ベクトルrow vectorsと呼ばれる。$A$の列から作られる$n$個の$\mathbb{R}^{m}$ベクターは

$$ \mathbf{c}_{1} = \begin{bmatrix} a_{11} \\ a_{21} \\ \vdots \\ a_{m1} \end{bmatrix},\quad \mathbf{c}_{2} = \begin{bmatrix} a_{12} \\ a_{22} \\ \vdots \\ a_{m2} \end{bmatrix},\quad \dots,\quad \mathbf{c}_{n} = \begin{bmatrix} a_{1n} \\ a_{2n} \\ \vdots \\ a_{mn} \end{bmatrix} $$

$A$の列ベクトルcolumn vectorsと呼ばれる。

$$ \begin{align*} A =& \begin{bmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn} \end{bmatrix} = \begin{bmatrix} \mathbf{r}_{1} \\ \mathbf{r}_{2} \\ \vdots \\ \mathbf{r}_{m} \end{bmatrix} \\ =& \begin{bmatrix} \mathbf{c}_{1} & \mathbf{c}_{2} & \cdots & \mathbf{c}_{n} \end{bmatrix} \end{align*} $$

$A$の行ベクトル$\mathbf{r}_{1}, \mathbf{r}_{2},\dots,\mathbf{r}_{m}$によって生成される$\mathbb{R}^{n}$の部分空間を$A$の行空間row spaceと言い、以下のように表記する。

$$ \mathcal{R} (A) \quad \text{or} \quad \text{row}(A) $$

$A$の列ベクトル$\mathbf{c}_{1}, \mathbf{c}_{2},\dots,\mathbf{c}_{n}$によって生成される$\mathbb{R}^{m}$の部分空間を$A$の列空間column spaceと言い、以下のように表記する。

$$ \mathcal{C} (A) \quad \text{or} \quad \text{col}(A) $$

同次連立一次方程式 $A \mathbf{x} =\mathbf{0}$の解集合を$A$の零空間null spaceと言い、以下のように表記する。

$$ \mathcal{N}(A) \quad \text{or} \quad \text{null}(A) $$

説明

上記の概念は

$$ \begin{equation} A\mathbf{x} = \mathbf{b} \end{equation} $$

連立一次方程式を解くために考案された。つまり、線形代数学では$(1)$の解と$A$の行空間、列空間、零空間の関係に興味があるのだ。具体的には行空間の基底を求めることが線形システムを解くことに関連している。特に、行空間と列空間の次元ランクと言い、零空間の次元を無効次元と言う。

なお、列空間は$\operatorname{Im} (A)$、すなわちimageとも呼ばれる。行列$A$を関数の概念として考えるならば、$A \in \mathbb{R}^{m \times n}$に対応する関数$T_{A}$は$T_{A} : \mathbb{R}^{n} \to \mathbb{R}^{m}$としても見ることができるからだ。

定理1

線形システム$A \mathbf{x} = \mathbf{b}$が解を持つための必要十分条件は$\mathbf{b} \in \mathcal{C}(A)$である。

定理2

$\mathbf{x}_{0}$が$A\mathbf{x} = \mathbf{b}$のある解だとしよう。$S= \left\{ \mathbf{v}_{1}, \mathbf{v}_{2}, \dots, \mathbf{v}_{k} \right\}$を$\mathcal{N}(A)$の基底としよう。そうすると、$A\mathbf{x} = \mathbf{b}$の全ての解は下記のように表現できる。

$$ \begin{equation} \mathbf{x} = \mathbf{x}_{0} + c_{1}\mathbf{v}_{1} + c_{2}\mathbf{v}_{2} + \cdots + c_{k}\mathbf{v}_{k} \end{equation} $$

逆に、全ての定数$c_{1}, c_{2}, \dots, c_{k}$に対して、上記の$\mathbf{x}$は$A\mathbf{x} = \mathbf{b}$の解である。


$(2)$を$A \mathbf{x} = \mathbf{b}$の一般解general solutionという。$\mathbf{x}_{0}$を$A \mathbf{x} = \mathbf{b}$の特殊解particular solutionという。そして、$c_{1}\mathbf{v}_{1} + c_{2}\mathbf{v}_{2} + \cdots + c_{k}\mathbf{v}_{k}$を$A \mathbf{x} = \mathbf{0}$の一般解という。

これらの定理から、非同次線形システムの一般解はある特殊解と同次線形システムの一般解の和として表されることが分かる。

参照

抽象代数学における核

零空間は$\ker A$と書き、kernelとも呼ばれる。これは抽象代数学で扱われる一般的な核の概念を線形代数学で特殊化した表現であり、これも$A$を関数と見なしたものだ。


  1. Howard Anton, Chris Rorres, Anton Kaul, Elementary Linear Algebra: Applications Version(12th Edition). 2019, p263-267 ↩︎