Skip to content

Commit

Permalink
docs: 完善第 2, 3 讲例题答案和定理证明 (yhwu-is#30)
Browse files Browse the repository at this point in the history
* docs: 更新内积空间上的算子.

* docs: update 第 25 讲.

* docs: update lec 25 & 27.

* docs: 完善第 2,3 讲例题解答及定理证明

---------

Co-authored-by: 45gfg9 <[email protected]>
  • Loading branch information
shadowash0215 and 45gfg9 authored Sep 18, 2023
1 parent 0f3ccb6 commit e389b39
Show file tree
Hide file tree
Showing 3 changed files with 154 additions and 27 deletions.
29 changes: 29 additions & 0 deletions 讲义/专题/2 线性空间.tex
Original file line number Diff line number Diff line change
Expand Up @@ -117,6 +117,35 @@ \section{线性子空间}
\end{enumerate}
\end{example}

\begin{solution}
\begin{enumerate}
\item 只需证明$\mathbf{R}[x]_2 \subseteq \mathbf{R}[x]_3$,以及$\mathbf{R}[x]_2$$\mathbf{R}[x]_3$中的加法和数乘封闭即可.

$\forall v \in \mathbf{R}[x]_2$,可被写作$v=a+bx,a,b \in \mathbf{R}$. 又有$\mathbf{R}[x]_3=\{a+bx+cx^2,a,b,c \in \mathbf{R}\}$,取$c=0$,有$v=a+bx \in \mathbf{R}[x]_3$,因此$\mathbf{R}[x]_2 \subseteq \mathbf{R}[x]_3$.

对于$\mathbf{R}[x]_3$中的加法和数乘:
\[mv_1+nv_2=m(a_1+b_1x)+n(a_2+b_2x)=(ma_1+na_2)+(mb_1+nb_2)x \in \mathbf{R}[x]_3\]
所以$\mathbf{R}[x]_2$$\mathbf{R}[x]_3$的子空间.

\item$W_1$: 引入参数$t$
\[W_1=\left\{(3t,2t,t) \,\middle|\, \frac{x}{3} = \frac{y}{2} = z = t\right\}\]
对于$\forall v_1, v_2 \in W_1, v_1 = (3t_1, 2t_1, t_1), v_2 = (3t_2, 2t_2, t_2)$,有
\begin{align*}
av_1 + bv_2 & = (3at_1 + 3bt_2, 2at_1 + 2bt_2, at_1 + bt_2) \\
& = (3(at_1 + bt_2), 2(at_1 + bt_2), at_1 + bt_2) \in W_1
\end{align*}
$W_1$封闭,是 $\mathbf{R}^3$ 的子空间.

$W_2$: 有反例. 取 $u_1 = (1, 0, 0), u_2 = (0, 0, 1) \in W_2$,但 $W_1 + W_2 = (1, 0, 1)$ 不满足 $x + y + z = 1$,故 $W_2$ 不封闭,不是 $\mathbf{R}^3$ 的子空间.

\item 设齐次线性方程组 $AX=0$ 的解构成的集合是 $W_1$$\forall X_1, X_2 \in W_1$,有 $AX_1 = AX_2 = 0$,所以 $\forall a, b \in \mathbf{F}$
\[A(a X_1 + b X_2) = A(a X_1) + A(b X_2) = a AX_1 + b AX_2 = 0\]
$W_1$ 封闭,是 $\mathbf{F}^n$ 的子空间.

设非齐次线性方程组 $AX = \beta,\enspace \beta \in \mathbf{F}^m,\enspace \beta \neq 0$ 的解构成的集合是 $W_2$$\forall X_1, X_2 \in W_2$,有 $AX_1 = AX_2 = \beta$,所以 $A(X_1 + X_2) = AX_1 + AX_2 = 2\beta \neq \beta$. 故 $W_2$ 不封闭,不是 $\mathbf{F}^n$ 的子空间.
\end{enumerate}
\end{solution}

上例中 \ref*{item:2:常见子空间:2} 表明过原点的直线/平面构成三维空间的子空间,不过原点的无法保持线性性. 事实上 \ref*{item:2:常见子空间:2} 和 \ref*{item:2:常见子空间:3} 在表述同一个问题,\ref*{item:2:常见子空间:2} 从几何角度描述了 \ref*{item:2:常见子空间:3} 中齐次/非齐次线性方程组的解集. 事实上,在定义了子空间后, 如果一个线性空间的子集也构成线性空间,我们就可以对其进行同样的研究. 这一想法在我们后续的内容中十分重要, 现在需要大家先熟知子空间的定义和判别.

最后我们需要注意一个名词的定义. 线性空间有两个子空间称为平凡子空间,即仅含零元的子集$\{0\}$和其自身$V$. 而其它子空间称为非平凡子空间.
Expand Down
2 changes: 1 addition & 1 deletion 讲义/专题/25 内积空间上的算子(II).tex
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ \section{正交矩阵和酉矩阵}

本节我们将唤醒一些沉睡的记忆,如果你已经忘了过渡矩阵或矩阵的相似,可以移步到前面的章节再回顾一下. 如果你还在这的话,那么坐稳,我们马上开始.

\subsection{定义 \quad Schur 定理}
\subsection{定义}

为了更好地引进正交矩阵和酉矩阵,我们有必要把共轭转置说的更清楚些. 共轭转置有着以下的运算性质,虽然都是看起来很显然的事情,此处还是稍稍赘述一下:

Expand Down
Loading

0 comments on commit e389b39

Please sign in to comment.