主要内容

高等代数教学辅导

4.2 线性映射和运算

建设中!

子节 4.2.1 主要知识点

定义 4.2.1.

\(U\)\(V\)是数域\(\mathbb{F} \)上的两个线性空间。若映射\(\phi: V\to U\)满足
  1. 对任意的\(\alpha_1\)\(\alpha_2\in V\),总有
    \begin{equation*} \phi(\alpha_1+\alpha_2) = \phi(\alpha_1)+\phi(\alpha_2) \end{equation*}
  2. 对任意的\(c\in \mathbb{F}\)\(\alpha\in V\),总有
    \begin{equation*} \phi(c\alpha) = c\phi(\alpha) \end{equation*}
则称\(\phi\)\(V\)\(U\)线性映射 。 特别的, 当\(V=U\)时,称\(\phi\)线性变换 ; 当\(U =\mathbb{F}\) 时, 称\(\phi\)线性函数

4.2.2.

\(V\)是数域\(\mathbb{F}\)上的\(n\)维线性空间,\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)\(V\)的一组基,定义
\begin{equation*} \phi: V\to \mathbb{F}^n, \alpha\mapsto (a_1,a_2,\ldots,a_n)^T, \end{equation*}
其中\((a_1,a_2,\ldots,a_n)^T\)\(\alpha\)在基\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)下的坐标,则\(\phi\)\(V\)\(\mathbb{F}^n\)的线性映射。

4.2.3.

线性空间\(V\)\(U\)的零映射\(0:\alpha\mapsto 0\)是线性映射。

4.2.4.

线性空间\(V\)的恒等映射\({\rm id}_V\)\(V\)的线性变换。

4.2.5.

\(V_1\)\(V_2\)\(V\)的子空间,\(V=V_1\oplus V_2\)。对\(i =1\)\(2\) 定义
\begin{gather*} \tau_i: V\to V_i,\ \alpha_1+\alpha_2\mapsto \alpha_i,\\ \sigma_1: V_1\to V,\ \alpha_1\mapsto \alpha_1+0,\\ \sigma_2: V_2\to V,\ \alpha_2\mapsto 0+\alpha_2, \end{gather*}
\(\tau_i\)\(\sigma_i\)是线性映射,称\(\tau_i\)投影映射\(\sigma_{i}\)嵌入映射

4.2.6.

\(A\in \mathbb{F}^{m\times n}\)。 验证如下定义的映射
\begin{equation*} \phi: \mathbb{F}^n\to \mathbb{F}^m,\ X\mapsto AX \end{equation*}
是线性映射。
  • \(r(A) = n \Leftrightarrow \phi\)为单线性映射;
  • \(r(A) = m \Leftrightarrow \phi\)为满线性映射。

4.2.7.

\(V=\mathbb{R}^2\)(实数域上二维向量空间),用\(T_\theta\)表示把\(V\)中每一个向量绕坐标原点旋转\(\theta\)角的变换,则\(T_\theta\)是一个线性变换。即
\begin{equation*} T_\theta: \mathbb{R}^2\to \mathbb{R}^2,\ \begin{pmatrix} x\\y \end{pmatrix} \mapsto \begin{pmatrix} x'\\y' \end{pmatrix} \end{equation*}
这里
\begin{equation*} \begin{pmatrix} x'\\y' \end{pmatrix} = \begin{pmatrix} \cos\theta & -\sin\theta\\ \sin\theta & \cos\theta \end{pmatrix} \begin{pmatrix} x\\y \end{pmatrix}. \end{equation*}
  • 线性映射 \(\phi\) 是单射 \(\Leftrightarrow\)\(\phi(\alpha)=0\),则 \(\alpha=0\)
  • 线性映射保持线性组合及关系式不变,即
    \begin{equation*} \phi(\sum_{i=1}^s k_i \alpha_i) =\sum_{i=1}^sk_i\phi(\alpha_i). \end{equation*}
  • 可作为 \(\phi\) 是线性映射的等价定义。
  1. 线性映射保零元、保负元、保线性组合。
  2. 线性映射保线性相关,但不保线性无关。单的线性映射保线性无关。

定义 4.2.11.

\(\phi\)\(\psi\)\(\mathbb{F}\)上线性空间\(V\to U\)的线性映射,则\(\phi\)\(\psi\) \(\phi+\psi\)定义为
\begin{equation*} \phi+\psi: V\to U,\ \alpha\mapsto\phi(\alpha)+\psi(\alpha) \end{equation*}
\(\phi+\psi\)也是\(V\to U\)的线性映射。
  • \(-\phi\)也称为\(\phi\)负映射

定义 4.2.13.

\(\phi\)\(\mathbb{F}\)上线性空间\(V\)\(U\)的线性映射, \(c\)\(\mathbb{F}\)的任意数,则\(c\)\(\phi\)数乘\(c\phi\)定义为
\begin{equation*} c\phi: V\to U,\ \alpha\mapsto c\phi(\alpha) \end{equation*}
\(c\phi\)也是\(V\to U\) 的线性映射。
\(V\)\(U\)\(\mathbb{F}\)上线性空间,\(\phi\in \mathcal{L}(V,U)\),分别取\(V\)\(U\)的基为:
\begin{equation*} V: \varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n;\quad U: \eta_1,\eta_2,\ldots,\eta_m. \end{equation*}
\(\begin{cases} \phi(\varepsilon_1)= {\color{red}a_{11}}\eta_1+{\color{red}a_{21}}\eta_2+{\color{red}\cdots}+{\color{red}a_{m1}}\eta_m\\ \phi(\varepsilon_2)= {\color{blue}a_{12}}\eta_1+{\color{blue}a_{22}}\eta_2+{\color{blue}\cdots}+{\color{blue}a_{m2}}\eta_m\\ {\hspace{3cm} \color{orange}\vdots}\\ \phi(\varepsilon_n)= {\color{green}a_{1n}}\eta_1+{\color{green}a_{2n}}\eta_2+{\color{green}\cdots}+{\color{green}a_{mn}}\eta_m \end{cases}\), 记
\begin{equation*} A =\begin{pmatrix} {\color{red}a_{11}} & {\color{blue}a_{12}} & {\color{orange}\cdots} & {\color{green}a_{1n}}\\ {\color{red}a_{21}} & {\color{blue}a_{22}} & {\color{orange}\cdots} & {\color{green}a_{2n}}\\ {\color{red}\vdots} & {\color{blue}{\vdots}} & {\color{orange}\ddots} & {\color{green}\vdots}\\ {\color{red}a_{m1}} & {\color{blue}a_{m2}} & {\color{orange}\cdots} & {\color{green}a_{mn}} \end{pmatrix}_{m\times n} \end{equation*}
则称矩阵\(A\)\(\phi\)\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵或表示矩阵 。形式上记为
\begin{equation*} {\color{red}\phi(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)=(\eta_1,\eta_2,\ldots,\eta_m)A}. \end{equation*}

定义 4.2.17.

\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)为数域\(\mathbb{F}\)上线性空间\(V\)的一组基,\(\phi\)\(V\)上的线性变换。设
\begin{equation*} \phi(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)=(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)A_{n\times n}. \end{equation*}
矩阵\(A\)称为线性变换\(\phi\)在基\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)下的矩阵。

练习 4.2.2 练习

1.

判断下列映射是否是线性映射,并说明理由。
  1. \(\varphi:\mathbb{F}^3\rightarrow\mathbb{F}^2,\ (x,y,z)^T\mapsto (x-2z,y-z)^T\text{;}\)
  2. \(\varphi:\mathbb{F}^3\rightarrow\mathbb{F}^3,\ (x,y,z)^T\mapsto (x^2,y^2,z^2)^T\text{;}\)
  3. \(\varphi:\mathbb{F}^2\rightarrow\mathbb{F}^2,\ (x,y)^T\mapsto (x+a,y+b)^T\),其中\((a,b)^T\)\(\mathbb{F}^2\)中固定的向量;
  4. \(\varphi:\mathbb{F}^{m\times n}\rightarrow\mathbb{F}^{n\times m},\ X\mapsto X^T\text{;}\)
  5. \(\varphi:\mathbb{F}^{s\times t}\rightarrow\mathbb{F}^{m\times n},\ X\mapsto AXB\),其中\(A\in\mathbb{F}^{m\times s},B\in\mathbb{F}^{t\times n}\)是固定的矩阵;
  6. \(\varphi:\mathbb{F}^{n\times n}\rightarrow\mathbb{F}^{n\times n},\ X\mapsto XAX\),其中\(A\in\mathbb{F}^{n\times n}\)是固定的一个矩阵;
  7. \(\varphi :\mathbb{R}\rightarrow\mathbb{R}^+,\ x\mapsto 5^x\),其中\(\mathbb{R}^+\)作为\(\mathbb{R}\)上的线性空间,加法、数乘运算分别定义为\(a\oplus b=ab,\ k \odot a = a^k,\ \forall a,b\in\mathbb{R}^+,k\in\mathbb{R}\)。。
解答.
  1. 对任意\((x_1,y_1,z_1)^T,(x_2,y_2,z_2)^T\in\mathbb{F}^3,c_1,c_1\in\mathbb{F}\),有
    \begin{equation*} \begin{array}{cl}&\varphi\left(c_1(x_1,y_1,z_1)^T+c_2(x_2,y_2,z_2)^T\right)\\=&\left((c_1x_1+c_2x_2)-2(c_1z_1+c_2z_2),(c_1y_1+c_2y_2)-(c_1z_1+c_2z_1)^T\right)\\ =&c_1(x_1-2z_1,y_1-z_1)^T+c_2(x_2-2z_2,y_2-z_2)^T\\ =&c_1\varphi\left((x_1,y_1,z_1)^T\right)+c_2\varphi\left((x_2,y_2,z_2)^T\right),\end{array} \end{equation*}
    所以\(\varphi\)是线性映射。
  2. 因为
    \begin{equation*} \varphi\left(2(1,0,0)^T\right)=(4,0,0)^T\neq (2,0,0)^T=2\varphi\left((1,0,0)^T\right), \end{equation*}
    所以\(\varphi\)不是线性映射。
  3. \((a,b)^T=(0,0)^T\)时,\(\varphi={\rm id}_{\mathbb{F}^2}\)。此时\(\varphi\)是线性映射。
    \((a,b)^T\neq (0,0)^T\)时,
    \begin{align*} \varphi\left(2(x,y)^T\right)& = &(2x+a,2y+b)^T \\ & = &2\varphi\left((x,y)^T\right) \end{align*}
    此时\(\varphi\)不是线性映射。
  4. 对任意\(X_1,X_2\in\mathbb{F}^{m\times n},c_1,c_2\in\mathbb{F}\),有
    \begin{align*} \varphi(c_1X_1+c_2X_2) & = &(c_1X_1+c_2X_2)^T\\ & = &c_1X_1^T+c_2X_2^T\\ & = &c_1\varphi(X_1)+c_2\varphi(X_2), \end{align*}
    所以\(\varphi\)是线性映射。
  5. 对任意\(X_1,X_2\in\mathbb{F}^{m\times n},c_1,c_2\in\mathbb{F}\),有
    \begin{align*} \varphi(c_1X_1+c_2X_2) & = &A(c_1X_1+c_2X_2)B\\ & = &c_1(AX_1B)+c_2(AX_2B)\\ & = & c_1\varphi(X_1)+c_2\varphi(X_2), \end{align*}
    所以\(\varphi\)是线性映射。
  6. \(A=0\)时,\(\varphi=0\)是线性映射。
    \(A\neq 0\)时,存在\(1\leq i,j\leq n\)使得\(a_{ij}\neq 0\),则
    \begin{align*} \varphi(2E_{ji})& = &(2E_{ji})A(2E_{ji})=4a_{ij}E_{ji}\\ &{\color{red} \neq} &2a_{ij}E_{ji}=E_{ji}A2E_{ji}=2\varphi(E_{ji}), \end{align*}
    此时\(\varphi\)不是线性映射。
  7. 对任意\(a,b\in\mathbb{R},k\in\mathbb{R}\),因为
    \begin{equation*} \varphi(a+b)=5^{a+b}=5^a\cdot 5^b=5^a\oplus 5^b=\varphi(a)\oplus \varphi(b), \end{equation*}
    \begin{equation*} \varphi(ka)=5^{ka}=(5^a)^k=k\odot 5^a=k\odot\varphi(a), \end{equation*}
    所以\(\varphi\)是线性映射。

2.

\(A\in\mathbb{F}^{m\times n},B\in\mathbb{F}^{m\times s}\)是两个固定的矩阵,证明:映射
\begin{equation*} \varphi:\mathbb{F}^{n\times s}\rightarrow\mathbb{F}^{m\times s},\ X\mapsto AX+B \end{equation*}
是线性映射的充要条件为\(B=0\)
解答.
充分性:由于\(B=0\),所以对任意\(X_1,X_2\in\mathbb{F}^{n\times s},k_1,k_2\in\mathbb{F}\),有
\begin{equation*} \varphi (k_1 X_1+k_2 X_2)=A(k_1 X_1+k_2 X_2)=k_1 (AX_1)+k_2 (AX_2)=k_1\varphi(X_1)+k_2 \varphi(X_2). \end{equation*}
\(\varphi\)\(\mathbb{F}^n\)上的线性变换。
必要性:由\(\varphi\)是线性映射知\(\varphi (X_1+X_2)=\varphi (X_1)+\varphi (X_2),\)
\begin{equation*} A(X_1+X_2)+B=(A X_1+B)+(AX_2+B), \end{equation*}
因此\(B=0\)

3.

\(\xi_1,\xi_2,\ldots,\xi_n\)是线性空间\(V\)的一个基,\(\varphi\)\(V\)\(U\)的线性映射。证明:\(\varphi\)是可逆映射的充要条件是\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)\(U\)的一个基。
解答.
只需证明\(\varphi\)是双射\(\Leftrightarrow\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)\(U\)的一个基。
\(\Rightarrow\)因为\(\varphi\)是单射,\(\varphi\)保持线性无关性,所以\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)线性无关。
\(\forall \beta\in U\),因为\(\varphi\)是满射,所以存在\(\alpha\in V\),使得\(\beta=\varphi(\alpha)\)。因为\(\xi_1,\xi_2,\ldots,\xi_n\)是基,所以\(\alpha=\sum_{i=1}^n c_i\xi_i\),于是
\begin{equation*} \beta=\varphi(\alpha)=\varphi(\sum_{i=1}^n c_i\xi_i)=\sum_{i=1}^n c_i\varphi(\xi_i). \end{equation*}
综上,\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)\(U\)的基。
\(\Leftarrow\)\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)\(U\)的基。对\(\forall \beta\in U\),存在\(c_1,c_2,\cdots ,c_n\)使得 \(\beta=c_1\varphi(\xi_1)+c_2\varphi(\xi_2)+\ldots+c_n\varphi(\xi_n)=\varphi(c_1\xi_1+c_2\xi_2+\ldots +c_n\xi_n),\)\(\varphi\)是满射。
以下只需证明\(\varphi(\alpha)=0\Rightarrow \alpha=0\)。设\(\alpha=\sum\limits_{i=1}^n c_i\xi_i\)\(\varphi(\alpha)=0\),则\(\sum\limits_{i=1}^n c_i\varphi(\xi_i)=0\)。由于\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)是基,所以\(c_1=c_2=\cdots=c_n=0\)。故\(\alpha=0\)

4.

\(V\)\(U\)是数域\(\mathbb{F}\)上的线性空间。\(\varphi\)\(V\to U\)的线性映射,\(V_1\)\(V_2\)\(V\)的子空间,\(U_1\)\(U\) 的子空间,\(S\)\(V\)的子集,证明:
  1. \(\varphi(V_1)\)\(U\)的子空间,\(\varphi^{-1}(U_1)\)\(V\)的子空间;
  2. \(\varphi(V_1+V_2)\)等于 \(\varphi(V_1)+\varphi(V_2)\)
  3. \(\varphi(\langle S\rangle)\)等于 \(\langle\varphi(S)\rangle\)
  4. 举例说明\(\varphi(V_1\bigcap V_2)\)未必等于 \(\varphi(V_1)\bigcap\varphi(V_2)\)
  5. 举例说明\(V_1\oplus V_2\not\Rightarrow\varphi(V_1)\oplus\varphi(V_2)\)
解答.
  1. \(\forall \tilde{\alpha},\tilde{\beta}\in \varphi(V_1)\)\(\exists \alpha,\beta\in V_1\),使得\(\varphi(\alpha)=\tilde{\alpha}\)\(\varphi(\beta)=\tilde{\beta}\)
    因为\(V_1\)\(V\)的子空间,所以对\(c_1c_2\in\mathbb{F}\)\(c_1 \alpha+c_2 \beta\in V_1\),于是
    \begin{equation*} c_1\tilde{\alpha}+c_2\tilde{\beta}= c_1\varphi(\alpha)+c_2\varphi(\beta)=\varphi(c_1 \alpha+c_2 \beta)\in\varphi(V_1) \end{equation*}
    所以\(\varphi(V_1)\)对线性组合封闭,推出\(\varphi(V_1)\)\(U\)的子空间。
    \(\forall \alpha,\beta\in \varphi^{-1}(U_1)\),有\(\varphi(\alpha),\varphi(\beta)\in U_1\)。因\(U_1\)是子空间,对\(\forall c_1,c_2\in \mathbb{F}\)
    \begin{equation*} \varphi(c_1 \alpha+c_2 \beta)=c_1\varphi(\alpha)+c_2\varphi(\beta)\in U_1 \end{equation*}
    于是\(c_1 \alpha+c_2 \beta\in \varphi^{-1}(U_1) \)\(\varphi^{-1}(U_1)\)对线性组合封闭\(\Rightarrow \varphi^{-1}(U_1)\le V\)
  2. 先证明\(\varphi(V_1+V_2)\subseteq \varphi(V_1)+\varphi(V_2)\)
    \(\forall \alpha\in V_1+V_2\)\(\exists \alpha_1\in V_1,\alpha_2\in V_2\), 使得\(\alpha=\alpha_1+\alpha_2\)。于是
    \begin{equation*} \varphi(\alpha)=\varphi(\alpha_1)+\varphi(\alpha_2)\in\varphi(V_1)+\varphi(V_2). \end{equation*}
    再证明\(\varphi(V_1)+\varphi(V_2)\subseteq \varphi(V_1+V_2)\)
    \(V_1\subseteq V_1+V_2,V_2\subseteq V_1+V_2\Rightarrow\varphi(V_1)\subseteq \varphi(V_1+V_2),\varphi(V_2)\subseteq \varphi(V_1+V_2)\),于是\(\varphi(V_1)+\varphi(V_2)\subseteq \varphi(V_1+V_2)\)
  3. 先证明\(\varphi(\langle S\rangle)\subseteq\langle\varphi(S)\rangle\)。 对\(\forall \tilde{\alpha}\in \varphi(\langle S\rangle)\),存在\(\alpha=\sum_{i=1}^n c_is_i\),其中\(c_i\in\mathbb{F},s_i\in S\),使得\(\varphi(\alpha)=\tilde{\alpha}\)。于是
    \begin{equation*} \tilde{\alpha}=\sum_{i=1}^nc_i\varphi(s_i)\in\langle\varphi(S)\rangle \end{equation*}
    再证明\(\langle\varphi(S)\rangle\subseteq\varphi(\langle S\rangle)\)。对\(\forall \tilde{\beta}\in \langle\varphi(S)\rangle\)
    \begin{equation*} \tilde{\beta}=\sum_{j=1}^m d_j\varphi(s_j)=\varphi(\sum_{j=1}^md_js_j)\in \varphi(\langle S\rangle). \end{equation*}
  4. \(V=\mathbb{F}^2\)\(V_1=\{(x,0)^T|x\in \mathbb{F}\}\)\(V_2=\{(0,y)^T|y\in\mathbb{F}\}\)
    \begin{equation*} \varphi: V\to \mathbb{F},\ (x,y)^T\mapsto x+y, \end{equation*}
    \(V_1\cap V_2=\{0\}\),但\(\varphi(V_1)\cap\varphi(V_2)=\mathbb{F}\)
  5. 同上,\(V_1\cap V_2=\{0\}\),但\(\varphi(V_1)\cap\varphi(V_2)=\mathbb{F}\neq \{0\}\)

5.

\(V\)是数域\(\mathbb{F}\)上线性空间,\(\varphi :V\rightarrow V\)是线性变换,\(V_1\)\(V_2\)\(V\)的子空间,且\(V=V_1\oplus V_2\)。证明:\(\varphi\)是可逆映射的充分必要条件是
\begin{equation*} V=\varphi(V_1)\oplus \varphi(V_2). \end{equation*}
解答.
必要性:因为\(\varphi\)是满射,所以\(V=\varphi(V)\)。又\(V=V_1+V_2\),故
\begin{equation*} V=\varphi(V_1+V_2)=\varphi(V_1)+\varphi(V_2). \end{equation*}
对任意\(\beta\in\varphi(V_1)\bigcap \varphi(V_2)\),存在\(\alpha_1\in V_1,\alpha_2\in V_2\)使得\(\beta=\varphi(\alpha_1),\beta=\varphi(\alpha_2)\)。由\(\varphi\)是单射知\(\alpha_1=\alpha_2\in V_1\bigcap V_2\),注意到\(V_1+V_2\)是直和,所以\(\alpha_1=\alpha_2=0\),于是\(\beta=\varphi(\alpha_1)=0\),因此\(\varphi(V_1)\bigcap \varphi(V_2)=\{0\}\)。综上,\(V =\varphi(V_1)\oplus \varphi(V_2)\)
充分性:由\(V=V_1+V_2\)\(\varphi(V)=\varphi(V_1)+\varphi(V_2)\),又\(V=\varphi(V_1)+\varphi(V_2)\),故\(V=\varphi(V)\),即\(\varphi\)是满射。下证\(\varphi\)是单射,即证由\(\varphi(\alpha)=0\)可推出\(\alpha=0\)
证法一:设\(\xi_1,\cdots ,\xi_r\)\(V_1\)的一个基,\(\xi_{r+1},\cdots ,\xi_n\)\(V_2\)的一个基,则\(\xi_1,\cdots ,\xi_n\)\(V\)的一个基且
\begin{equation*} \varphi(V_1)=\langle\varphi(\xi_1),\cdots,\varphi(\xi_r)\rangle,\varphi(V_2)=\langle\varphi(\xi_{r+1}),\cdots,\varphi(\xi_n)\rangle. \end{equation*}
于是\(\dim\varphi(V_1)\leq r,\dim\varphi(V_2)\leq n-r\)。由\(V=\varphi(V_1)\oplus \varphi(V_2)\)\(\dim\varphi(V_1)+\dim\varphi(V_2)=n\),所以\(\dim\varphi(V_1)= r\)\(\dim\varphi(V_2)= n-r\)。于是\(\varphi(\xi_1),\cdots ,\varphi(\xi_r)\)\(\varphi(V_1)\)的一个基, \(\varphi(\xi_{r+1}),\cdots,\varphi(\xi_n)\)\(\varphi(V_2)\)的一个基,\(\varphi(\xi_1),\cdots ,\varphi(\xi_n)\)\(V\)的一个基。
\(\varphi(\alpha)=0\),由于\(\xi_1,\cdots ,\xi_n\)\(V\)的一个基,所以存在\(c_1,\cdots ,c_n\in\mathbb{F}\),使得\(\alpha=\sum\limits_{i=1}^n c_i \xi_i\)。于是
\begin{equation*} 0=\varphi(\alpha)=\sum\limits_{i=1}^n c_i \varphi(\xi_i), \end{equation*}
\(\varphi(\xi_1),\cdots ,\varphi(\xi_n)\)线性无关得
\begin{equation*} c_1=\cdots =c_n=0, \end{equation*}
\(\alpha=0\),即\(\varphi\)是单射。因此\(\varphi\)是可逆变换。
证法二:(反证法)假设存在非零向量\(\alpha\in V\),使得\(\varphi(\alpha)=0\)。由\(V=V_1\oplus V_2\)知:存在\(\alpha_1\in V_1,\alpha_2\in V_2\),使得\(\alpha=\alpha_1+\alpha_2\),这里\(\alpha_1,\alpha_2\)不全为零向量。不妨设\(\alpha_1\neq 0\),则可将其扩充为\(V_1\)的一个基\(\alpha_1,\beta_2,\cdots ,\beta_r\)。由4、(3)结论得
\begin{equation*} \varphi(V_1)=\varphi(\langle\alpha_1,\beta_2,\cdots ,\beta_r\rangle)=\langle\varphi(\alpha_1),\varphi(\beta_2)\cdots,\varphi(\beta_r)\rangle, \end{equation*}
\(\dim \varphi(V_1)\leq r=\dim V_1\)。同理,\(\dim\varphi(V_2)\leq\dim V_2\)。由于
\begin{equation*} 0=\varphi(\alpha)=\varphi(\alpha_1)+\varphi(\alpha_2), \end{equation*}
这里\(\varphi(\alpha_1)\in\varphi(V_1),\varphi(\alpha_2)\in\varphi(V_2)\),又\(\varphi(V_1)+\varphi(V_2)\)是直和,所以
\begin{equation*} \varphi(\alpha_1)=\varphi(\alpha_2)=0. \end{equation*}
\(\varphi(V_1)=\langle\varphi(\beta_2)\cdots,\varphi(\beta_r)\rangle\),即\(\dim\varphi (V_1)<r=\dim V_1\)。于是
\begin{equation*} \dim\varphi(V_1)+\dim\varphi(V_2)<\dim V_1+\dim V_2, \end{equation*}
这与\(V=V_1\oplus V_2\)\(V=\varphi(V_1)\oplus \varphi(V_2)\)相矛盾。因此\(\varphi\)是单射。从而\(\varphi\)是可逆变换。

6.

\(\varphi:\mathbb{F}^4\rightarrow\mathbb{F}^3, \begin{pmatrix} a_1\\a_2\\a_3\\a_4 \end{pmatrix}\mapsto \begin{pmatrix} -a_1+a_2+2a_3+a_4\\-2a_2+a_3\\-a_1-a_2+3a_3+a_4 \end{pmatrix}\)
  1. 证明:\(\varphi\)是线性映射;
  2. \(\varphi\)在标准基\(\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4\)\(\varepsilon_1,\varepsilon_2,\varepsilon_3\)下的矩阵;
  3. \(\mathbb{F}^4\)内取一个基
    \begin{equation*} \alpha_1=(1,0,1,1)^T,\alpha_2=(0,1,0,1)^T,\alpha_3=(0,0,1,0)^T,\alpha_4=(0,0,2,1)^T, \end{equation*}
    又在\(\mathbb{F}^3\)内取一个基
    \begin{equation*} \beta_1=(1,1,1)^T,\beta_2=(1,0,-1)^T,\beta_3=(0,1,0)^T, \end{equation*}
    \(\varphi\)在基\(\alpha_1,\alpha_2,\alpha_3,\alpha_4\)\(\beta_1,\beta_2,\beta_3\)下的矩阵。
解答.
  1. 对任意\(\alpha=\begin{pmatrix} a_1\\a_2\\a_3\\a_4 \end{pmatrix},\beta=\begin{pmatrix} b_1\\b_2\\b_3\\b_4 \end{pmatrix}\in\mathbb{F}^4,c,d\in\mathbb{F}\),有
    \begin{equation*} \begin{array}{ccl}\varphi(c \alpha+d \beta)&=&\begin{pmatrix} -(ca_1+db_1)+(ca_2+db_2)+2(ca_3+db_3)+(ca_4+db_4)\\-2(ca_2+db_2)+(ca_3+db_3)\\-(ca_1+db_1)-(ca_2+db_2)+3(ca_3+db_3)+(ca_4+db_4) \end{pmatrix}\\ &=&c\begin{pmatrix} -a_1+a_2+2a_3+a_4\\-2a_2+a_3\\-a_1-a_2+3a_3+a_4 \end{pmatrix}+d\begin{pmatrix} -b_1+b_2+2b_3+b_4\\-2b_2+b_3\\-b_1-b_2+3b_3+b_4 \end{pmatrix}\\ &=&c\varphi(\alpha)+d\varphi(\beta), \end{array} \end{equation*}
    所以\(\varphi\)是线性映射。
  2. 由于
    \begin{equation*} \begin{array}{l} \varphi(\varepsilon_1)=(-1,0,-1)^T=-\varepsilon_1-\varepsilon_3,\\ \varphi(\varepsilon_2)=(1,-2,-1)^T=\varepsilon_1-2 \varepsilon_2-\varepsilon_3,\\ \varphi(\varepsilon_3)=(2,1,3)^T=2\varepsilon_1+\varepsilon_2+3\varepsilon_3,\\ \varphi(\varepsilon_4)=(1,0,1)^T=\varepsilon_1+\varepsilon_3, \end{array} \end{equation*}
    \begin{equation*} \varphi (\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4)=(\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4)\begin{pmatrix} -1&1&2&1\\ 0&-2&1&0\\ -1&-1&3&1 \end{pmatrix}, \end{equation*}
    所以\(\varphi\)在标准基\(\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4\)\(\varepsilon_1,\varepsilon_2,\varepsilon_3\)下的矩阵为\(A=\begin{pmatrix} -1&1&2&1\\ 0&-2&1&0\\ -1&-1&3&1 \end{pmatrix}\text{.}\)
  3. 由于\((\alpha_1,\alpha_2,\alpha_3,\alpha_4)=(\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4)P,(\beta_1,\beta_2,\beta_3)=(\varepsilon_1,\varepsilon_2,\varepsilon_3)Q\),其中\(P=\begin{pmatrix} 1&0&0&0\\ 0&1&0&0\\ 1&0&1&2\\ 1&1&0&1 \end{pmatrix},Q=\begin{pmatrix} 1&1&0\\ 1&0&1\\ 1&-1&0 \end{pmatrix},\) 所以\(\varphi\)在基\(\alpha_1,\alpha_2,\alpha_3,\alpha_4\)\(\beta_1,\beta_2,\beta_3\)下的矩阵为
    \begin{equation*} Q^{-1}AP=\begin{pmatrix} \frac{5}{2}&1&\frac{5}{2}&6\\ -\frac{1}{2}&1&-\frac{1}{2}&-1\\ -\frac{3}{2}&-3&-\frac{3}{2}&-4 \end{pmatrix}. \end{equation*}

7.

两个函数
\begin{equation*} \xi_1=e^{ax}\cos bx,\quad \xi_2=e^{ax}\sin bx \end{equation*}
的所有实线性组合构成实数域上一个二维线性空间\(V\),求求导变换
\begin{equation*} \varphi:V\rightarrow V,f(x)\mapsto f'(x) \end{equation*}
在基\(\xi_1,\xi_2\)下的矩阵。
解答.
\begin{equation*} \because\varphi(\xi_1)=e^{ax}(a\cos bx-b\sin bx)=a \xi_1-b \xi_2=(\xi_1,\xi_2) \begin{pmatrix} a\\-b \end{pmatrix} \end{equation*}
\begin{equation*} \varphi(\xi_2)=e^{ax}(a\sin bx+b\cos bx)=b \xi_1+a \xi_2=(\xi_1,\xi_2) \begin{pmatrix} b\\a \end{pmatrix} \end{equation*}
所以
\begin{equation*} \varphi(\xi_1,\xi_2)=(\xi_1,\xi_2)\begin{pmatrix} a & b\\ -b & a \end{pmatrix}, \end{equation*}
\(\varphi\)在基\(\xi_1,\xi_2\)下的矩阵为\(\begin{pmatrix} a & b\\ -b & a \end{pmatrix}\)

8.

定义
\begin{equation*} \varphi_1:\mathbb{F}^{2\times 2}\to\mathbb{F}^{2\times 2},\ X\mapsto AX, \end{equation*}
\begin{equation*} \varphi_2:\mathbb{F}^{2\times 2}\to\mathbb{F}^{2\times 2},\ X\mapsto XB, \end{equation*}
\begin{equation*} \varphi_3: \mathbb{F}^{2\times 2} \to \mathbb{F}_{2\times 2},\ X\mapsto AXB, \end{equation*}
其中\(A=\begin{pmatrix} 0 & 1\\ -1 & 0 \end{pmatrix}, B=\begin{pmatrix} 1&2\\3&4 \end{pmatrix}\text{,}\)\(\varphi_1,\varphi_2,\varphi_3\)在基\(E_{11},E_{12},E_{21},E_{22}\)下的矩阵。
解答.
(1)因为
\begin{equation*} \begin{array}{l} \varphi_1(E_{11})=AE_{11}= \begin{pmatrix} 0&0\\-1&0 \end{pmatrix}=0E_{11}+0E_{12}-E_{21}+0E_{22},\\ \varphi_1(E_{12})=AE_{12}= \begin{pmatrix} 0&0\\0&-1 \end{pmatrix}=0E_{11}+0E_{12}+0E_{21}-E_{22},\\ \varphi_1(E_{21})=AE_{21}= \begin{pmatrix} 1&0\\0&0 \end{pmatrix}=E_{11}+0E_{12}+0E_{21}+0E_{22},\\ \varphi_1(E_{22})=AE_{22}= \begin{pmatrix} 0&1\\0&0 \end{pmatrix}=0E_{11}+E_{12}+0E_{21}+0E_{22},\\ \end{array} \end{equation*}
\begin{equation*} \varphi_1(E_{11},E_{12},E_{21},E_{22})=(E_{11},E_{12},E_{21},E_{22})\begin{pmatrix} 0&0&1&0\\ 0&0&0&1\\ -1&0&0&0\\ 0&-1&0&0 \end{pmatrix}, \end{equation*}
所以\(\varphi_1\)在基\(E_{11},E_{12},E_{21},E_{22}\)下的矩阵为\(\begin{pmatrix} 0&E_2\\ -E_2&0 \end{pmatrix}\)
(2)因为
\begin{equation*} \begin{array}{l} \varphi_2(E_{11})=E_{11}B= \begin{pmatrix} 1&2\\0&0 \end{pmatrix}=E_{11}+2E_{12}+0E_{21}+0E_{22},\\ \varphi_2(E_{12})=E_{12}B= \begin{pmatrix} 3&4\\0&0 \end{pmatrix}=3E_{11}+4E_{12}+0E_{21}+0E_{22},\\ \varphi_2(E_{21})=E_{21}B= \begin{pmatrix} 0&0\\1&2 \end{pmatrix}=0E_{11}+0E_{12}+E_{21}+2E_{22},\\ \varphi_2(E_{22})=E_{22}B= \begin{pmatrix} 0&0\\3&4 \end{pmatrix}=0E_{11}+0E_{12}+3E_{21}+4E_{22},\\ \end{array} \end{equation*}
\begin{equation*} \varphi_2(E_{11},E_{12},E_{21},E_{22})=(E_{11},E_{12},E_{21},E_{22})\begin{pmatrix} 1&3&0&0\\ 2&4&0&0\\ 0&0&1&3\\ 0&0&2&4 \end{pmatrix}, \end{equation*}
所以\(\varphi_2\)在基\(E_{11},E_{12},E_{21},E_{22}\)下的矩阵为\(\begin{pmatrix} 1&3&0&0\\ 2&4&0&0\\ 0&0&1&3\\ 0&0&2&4 \end{pmatrix}\)
(3)因为
\begin{equation*} \begin{array}{l} \varphi_3(E_{11})=AE_{11}B= \begin{pmatrix} 0&0\\-1&-2 \end{pmatrix}=0E_{11}+0E_{12}-E_{21}-2E_{22},\\ \varphi_3(E_{12})=AE_{12}B= \begin{pmatrix} 0&0\\-3&-4 \end{pmatrix}=0E_{11}+0E_{12}-3E_{21}-4E_{22},\\ \varphi_3(E_{21})=AE_{21}B= \begin{pmatrix} 1&2\\0&0 \end{pmatrix}=E_{11}+2E_{12}+0E_{21}+0E_{22},\\ \varphi_3(E_{22})=AE_{22}B= \begin{pmatrix} 3&4\\0&0 \end{pmatrix}=3E_{11}+4E_{12}+0E_{21}+0E_{22},\\ \end{array} \end{equation*}
\begin{equation*} \varphi_3(E_{11},E_{12},E_{21},E_{22})=(E_{11},E_{12},E_{21},E_{22})\begin{pmatrix} 0&0&1&3\\ 0&0&2&4\\ -1&-3&0&0\\ -2&-4&0&0 \end{pmatrix}, \end{equation*}
所以\(\varphi_2\)在基\(E_{11},E_{12},E_{21},E_{22}\)下的矩阵为\(\begin{pmatrix} 0&0&1&3\\ 0&0&2&4\\ -1&-3&0&0\\ -2&-4&0&0 \end{pmatrix}\)

9.

\(V\)是数域\(\mathbb{F}\)上的线性空间,\(\dim V=1\)\(\varphi\)\(V\)上的线性变换。证明:存在\(c\in\mathbb{F}\),使得对任意\(\alpha\in V\),都有\(\varphi(\alpha)=c\alpha\)
解答.
\(\because\dim V=1\),所以取\(0\ne \alpha_0\in V\)\(\alpha_0\)\(V\)的基。
注意到\(\varphi(\alpha_0)\in V\)\(\alpha_0\)\(V\)的基, 则存在唯一\(c\in \mathbb{F}\)使得\(\varphi(\alpha_0)=c \alpha_0\)
\(\forall \alpha\in V\),存在\(k\in\mathbb{F}\)使得\(\alpha=k \alpha_0\),所以
\begin{equation*} \varphi(\alpha)=\varphi(k \alpha_0)=k\varphi(\alpha_0)=k(c \alpha_0)=c(k \alpha_0)=c \alpha. \end{equation*}

10.

\(W\)是有限维线性空间\(V\)的子空间,\(\varphi\)\(W\)\(V\)的线性映射,证明:\(\varphi\)可扩充为\(V\)上的线性变换\(\psi\),即存在线性变换\(\psi:V\rightarrow V\)使得
\begin{equation*} \psi (\alpha)=\varphi(\alpha),\ \forall\alpha\in W. \end{equation*}
解答.
\(\xi_1,\cdots ,\xi_r\)\(W\)的一个基,将其扩充为\(V\)的一个基\(\xi_1,\cdots ,\xi_r,\xi_{r+1},\cdots ,\xi_n\),定义
\begin{equation*} \psi:V\rightarrow V,\ \sum\limits_{i=1}^n c_i \xi_i\mapsto\sum\limits_{i=1}^{\color{blue}{r}} c_i \varphi(\xi_i), \end{equation*}
不难验证\(\psi\)\(V\)上线性变换,且对\(\forall \alpha=a_1 \xi_1+\cdots a_r \xi_r\in W\),有
\begin{equation*} \psi (\alpha)=a_1\varphi(\xi_1)+\cdots +a_r\varphi(\xi_r)=\varphi(\alpha). \end{equation*}

11.

设线性映射\(\varphi:U\rightarrow V\)在基\(\alpha_1,\alpha_2\)\(\beta_1,\beta_2,\beta_3\)下的矩阵为
\begin{equation*} A=\begin{pmatrix} 2&0\\0&-2\\1&-1 \end{pmatrix}, \end{equation*}
\(U\)中向量\(\alpha\)在基\(\xi_1=\alpha_1,\xi_2=\alpha_1+\alpha_2\)下的坐标为\((2,3)^T\),求\(\varphi(\alpha)\)在基\(\eta_1=\beta_1+\beta_2,\eta_2=-2\beta_1-\beta_2,\eta_3=2\beta_2-\beta_3\)下的坐标。
解答.
因为
\begin{equation*} (\xi_1,\xi_2)=(\alpha_1,\alpha_2)P,(\eta_1,\eta_2,\eta_3)=(\beta_1,\beta_2,\beta_3)Q, \end{equation*}
其中\(P=\begin{pmatrix} 1&1\\0&1 \end{pmatrix},Q=\begin{pmatrix} 1&-2&0\\ 1&-1&2\\ 0&0&-1 \end{pmatrix}\),所以\(\varphi\)\(\xi_1,\xi_2\)\(\eta_1,\eta_2,\eta_3\)下矩阵为
\begin{equation*} B=Q^{-1}AP=\begin{pmatrix} 2&-6\\ 0&-4\\ -1&0 \end{pmatrix}, \end{equation*}
因此\(\varphi(\alpha)\)在基\(\eta_1,\eta_2,\eta_3\)下的坐标为\(B(2,3)^T=(-14,-12,-2)^T\)

12.

\(V,U,W\)是数域\(\mathbb{F}\)上的线性空间,\(\varphi\in\mathcal{L}(V,U),\psi\in\mathcal{L}(U,W)\)。证明:
  1. \(\psi\varphi\in\mathcal{L} (V,W)\)
  2. \(\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵为\(A\)\(\psi\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\zeta_1,\zeta_2,\ldots,\zeta_s\)下的矩阵为\(B\),则\(\psi\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\zeta_1,\zeta_2,\ldots,\zeta_s\)下的矩阵为\(BA\)
  3. \(\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵为\(A\),则\(\varphi\)是可逆映射的充要条件是\(A\)是可逆矩阵。此时,\(\varphi^{-1}\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\xi_1,\xi_2,\ldots,\xi_n\)下的矩阵为\(A^{-1}\)
解答.
  1. 因为\(\varphi: V\to U,\psi:U\to W\),所以\(\psi\varphi:V\to W\)
    \(\forall \alpha_1,\alpha_2\in V,c_1,c_2\in\mathbb{F}\)
    \begin{align*} \end{align*}
    所以\(\psi\varphi\)是线性映射,即\(\psi\varphi\in\mathcal{L}(V,W)\)
  2. \(\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵为\(A\),即
    \begin{equation} \varphi(\xi_1,\xi_2,\ldots,\xi_n)=(\eta_1,\eta_2,\ldots,\eta_m)A\tag{4.1} \end{equation}
    \(\psi\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\zeta_1,\zeta_2,\ldots,\zeta_s\)下的矩阵为\(B\),即
    \begin{equation} \psi(\eta_1,\eta_2,\ldots,\eta_m)=(\zeta_1,\zeta_2,\ldots,\zeta_s)B\tag{4.2} \end{equation}
    (4.1), (4.2)
    \begin{align*} \psi\varphi(\xi_1,\xi_2,\ldots,\xi_n) & = & \psi[\varphi(\xi_1,\xi_2,\ldots,\xi_n)]\\ & = & \psi[(\eta_1,\eta_2,\ldots,\eta_m)A]\\ & = & [\psi(\eta_1,\eta_2,\ldots,\eta_m)]A\\ & = & (\zeta_1,\zeta_2,\ldots,\zeta_s)BA \end{align*}
  3. 充分性:定义\(U\)\(V\)的线性映射\(\psi\)满足
    \begin{equation*} \psi (\eta_1,\eta_2,\ldots,\eta_m)=(\xi_1,\xi_2,\ldots,\xi_n)A^{-1}, \end{equation*}
    \begin{equation*} \begin{array}{ccl}\psi\varphi (\xi_1,\xi_2,\ldots,\xi_n)&=&\psi \left((\eta_1,\eta_2,\ldots,\eta_m)A\right)=\psi (\eta_1,\eta_2,\ldots,\eta_m)A\\&=&(\xi_1,\xi_2,\ldots,\xi_n)A^{-1}A= {\rm id}_V(\xi_1,\xi_2,\ldots,\xi_n)\end{array} \end{equation*}
    \begin{equation*} \begin{array}{ccl}\varphi\psi (\eta_1,\eta_2,\ldots,\eta_m)&=&\varphi \left((\xi_1,\xi_2,\ldots,\xi_n)A^{-1}\right)=\varphi (\xi_1,\xi_2,\ldots,\xi_n)A^{-1}\\&=&(\eta_1,\eta_2,\ldots,\eta_m)AA^{-1}= {\rm id}_U(\eta_1,\eta_2,\ldots,\eta_m)\end{array} \end{equation*}
    可知\(\psi\varphi={\rm id}_V,\varphi\psi={\rm id}_U\)。故\(\varphi\)是可逆映射。此时,\(\varphi^{-1}\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\xi_1,\xi_2,\ldots,\xi_n\)下的矩阵为\(A^{-1}\)
    必要性:因为\(\varphi\)可逆,所以存在\(\sigma\in\mathcal{L} (U,V)\)使得
    \begin{equation*} \sigma\varphi= {\rm id}_V,\ \varphi\sigma= {\rm id}_U. \end{equation*}
    \(\sigma\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\xi_1,\xi_2,\ldots,\xi_n\)下的矩阵为\(B\),即
    \begin{equation*} \sigma (\eta_1,\eta_2,\ldots,\eta_m)=(\xi_1,\xi_2,\ldots,\xi_n)B, \end{equation*}
    \((2)\)\(\sigma\varphi\)在基\(\xi_1,\xi_2,\cdots ,\xi_n\)下的矩阵为\(BA\)\(\varphi\sigma\)在基\(\eta_1,\eta_2,\cdots ,\eta_m\)下的矩阵为\(AB\)。注意到\(\sigma\varphi= {\rm id}_V,\ \varphi\sigma= {\rm id}_U\),故\(BA=E\)\(AB=E\)。因此\(A\)是可逆矩阵,且\(B=A^{-1}\)。此时,\(\varphi^{-1}\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\xi_1,\xi_2,\ldots,\xi_n\)下的矩阵为\(A^{-1}\)

13.

\(\varphi\)\(3\)维线性空间\(V\)上的线性变换,\(\xi_1,\xi_2,\xi_3\)\(V\)的一个基,
\begin{equation*} \varphi(\xi_1)=\xi_1,\varphi(\xi_2)=\xi_1+\xi_2,\varphi(\xi_3)=\xi_1+\xi_2+\xi_3, \end{equation*}
证明:\(\varphi\)是可逆映射,并求\(\varphi^{-1}\)
解答.
\(A=\begin{pmatrix} 1&1&1\\ 0&1&1\\ 0&0&1 \end{pmatrix}\),因为
\begin{equation*} \varphi(\xi_1,\xi_2,\xi_3)=(\xi_1,\xi_2,\xi_3)A, \end{equation*}
\(A\)可逆,\(A^{-1}=\begin{pmatrix} 1&-1&0\\ 0&1&-1\\ 0&0&1 \end{pmatrix}\),所以由12、(3)结论知:\(\varphi\)是可逆映射,且
\begin{equation*} \varphi^{-1}(\xi_1,\xi_2,\xi_3)=(\xi_1,\xi_2,\xi_3)A^{-1}, \end{equation*}
\begin{equation*} \varphi^{-1}:V\rightarrow V,\ c_1 \xi_1+c_2 \xi_2+c_3 \xi_3\mapsto (c_1-c_2) \xi_1+(c_2-c_3)\xi_2+c_3 \xi_3. \end{equation*}