Jephian Lin
    • Create new note
    • Create a note from template
      • Sharing URL Link copied
      • /edit
      • View mode
        • Edit mode
        • View mode
        • Book mode
        • Slide mode
        Edit mode View mode Book mode Slide mode
      • Customize slides
      • Note Permission
      • Read
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
      • Write
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
      • Engagement control Commenting, Suggest edit, Emoji Reply
    • Invite by email
      Invitee

      This note has no invitees

    • Publish Note

      Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note

      Your note will be visible on your profile and discoverable by anyone.
      Your note is now live.
      This note is visible on your profile and discoverable online.
      Everyone on the web can find and read all notes of this public team.
      See published notes
      Unpublish note
      Please check the box to agree to the Community Guidelines.
      View profile
    • Commenting
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
      • Everyone
    • Suggest edit
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
    • Emoji Reply
    • Enable
    • Versions and GitHub Sync
    • Note settings
    • Note Insights New
    • Engagement control
    • Transfer ownership
    • Delete this note
    • Save as template
    • Insert from template
    • Import from
      • Dropbox
      • Google Drive
      • Gist
      • Clipboard
    • Export to
      • Dropbox
      • Google Drive
      • Gist
    • Download
      • Markdown
      • HTML
      • Raw HTML
Menu Note settings Note Insights Versions and GitHub Sync Sharing URL Create Help
Create Create new note Create a note from template
Menu
Options
Engagement control Transfer ownership Delete this note
Import from
Dropbox Google Drive Gist Clipboard
Export to
Dropbox Google Drive Gist
Download
Markdown HTML Raw HTML
Back
Sharing URL Link copied
/edit
View mode
  • Edit mode
  • View mode
  • Book mode
  • Slide mode
Edit mode View mode Book mode Slide mode
Customize slides
Note Permission
Read
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
Write
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
Engagement control Commenting, Suggest edit, Emoji Reply
  • Invite by email
    Invitee

    This note has no invitees

  • Publish Note

    Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note

    Your note will be visible on your profile and discoverable by anyone.
    Your note is now live.
    This note is visible on your profile and discoverable online.
    Everyone on the web can find and read all notes of this public team.
    See published notes
    Unpublish note
    Please check the box to agree to the Community Guidelines.
    View profile
    Engagement control
    Commenting
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    • Everyone
    Suggest edit
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    Emoji Reply
    Enable
    Import from Dropbox Google Drive Gist Clipboard
       Owned this note    Owned this note      
    Published Linked with GitHub
    • Any changes
      Be notified of any changes
    • Mention me
      Be notified of mention me
    • Unsubscribe
    # 喬丹標準型 ![Creative Commons License](https://i.creativecommons.org/l/by/4.0/88x31.png) This work by Jephian Lin is licensed under a [Creative Commons Attribution 4.0 International License](http://creativecommons.org/licenses/by/4.0/). $\newcommand{\trans}{^\top} \newcommand{\adj}{^{\rm adj}} \newcommand{\cof}{^{\rm cof}} \newcommand{\inp}[2]{\left\langle#1,#2\right\rangle} \newcommand{\dunion}{\mathbin{\dot\cup}} \newcommand{\bzero}{\mathbf{0}} \newcommand{\bone}{\mathbf{1}} \newcommand{\ba}{\mathbf{a}} \newcommand{\bb}{\mathbf{b}} \newcommand{\bc}{\mathbf{c}} \newcommand{\bd}{\mathbf{d}} \newcommand{\be}{\mathbf{e}} \newcommand{\bh}{\mathbf{h}} \newcommand{\bp}{\mathbf{p}} \newcommand{\bq}{\mathbf{q}} \newcommand{\br}{\mathbf{r}} \newcommand{\bx}{\mathbf{x}} \newcommand{\by}{\mathbf{y}} \newcommand{\bz}{\mathbf{z}} \newcommand{\bu}{\mathbf{u}} \newcommand{\bv}{\mathbf{v}} \newcommand{\bw}{\mathbf{w}} \newcommand{\tr}{\operatorname{tr}} \newcommand{\nul}{\operatorname{null}} \newcommand{\rank}{\operatorname{rank}} %\newcommand{\ker}{\operatorname{ker}} \newcommand{\range}{\operatorname{range}} \newcommand{\Col}{\operatorname{Col}} \newcommand{\Row}{\operatorname{Row}} \newcommand{\spec}{\operatorname{spec}} \newcommand{\vspan}{\operatorname{span}} \newcommand{\Vol}{\operatorname{Vol}} \newcommand{\sgn}{\operatorname{sgn}} \newcommand{\idmap}{\operatorname{id}} \newcommand{\am}{\operatorname{am}} \newcommand{\gm}{\operatorname{gm}}$ ```python from lingeo import random_good_matrix ``` ## Main idea A **Jordan block** $J_{\lambda,m}$ is an $m\times m$ matrix $$ J_{\lambda,m} = \begin{bmatrix} \lambda & 1 & 0 & \cdots & 0 \\ 0 & \lambda & 1 & \ddots & \vdots \\ 0 & 0 & \lambda & \ddots & 0 \\ \vdots & ~ & \ddots & \ddots & 1 \\ 0 & \cdots & 0 & 0 & \lambda \end{bmatrix}. $$ A Jordan block $J_{\lambda,m}$ has only one eigenvalue $\lambda$ with $\am(\lambda) = m$ and $\gm(\lambda) = 1$. Therefore, it is not diagonalizable whenever $m\geq 2$. ##### Theorem (Jordan canonical form) For any square matrix $A$ over $\mathbb{C}$ there is a basis $\beta$ such that $[f_A]_\beta^\beta$ is a block diagonal matrix whose diagonal blocks are Jordan blocks. A **generalized eigenvector** of $A$ with respect to $\lambda$ is a nonzero vector $\bv$ such that $$ (A - \lambda I)^k\bv = \bzero $$ for some $k$. Equivalently, $p(A)\bv = \bzero$ for $p(x) = (x - \lambda)^k$. Therefore, $m_{A,\bv}(x) \mid (x - \lambda)^k$ and $p(x)$ can be chosen as $m_{A,\bv}(x) = (x - \lambda)^d$ for some $d \leq n$ instead. Consequently, the set of all generalized eigenvectors of $A$ with respect to $\lambda$ are exactly the nonzero vectors in $$ F_\lambda = \ker(A - \lambda I)^n, $$ and we call $F_\lambda$ as the **generalized eigenspace** of $A$ with respect to $\lambda$. A subspace $W$ is called an $A$-invariant subspace if $$ \{A\bw: \bw\in W\} \subseteq W. $$ If $W$ is an $A$-invariant subspace, then the restriction $f_A\big|_W: W \rightarrow W$ defined as $f_A(\bw) = A\bw$ is a well-defined function with $\spec(f_A\big|_W) \subseteq \spec(f_A)$. Indeed, if $\alpha$ is a basis of $W$ and $\beta\supseteq\alpha$ is an extendsion of $\alpha$ as a basis of $\mathbb {R}^n$, then $$ [f_A]_\beta^\beta = \begin{bmatrix} A_W & B \\ O & C \end{bmatrix}, $$ where $A_W = [f_A\big|_W]_\alpha^\alpha$. Therefore, $F_\lambda$ is an $A$-invariant subspace. A foundation of the theorem of Jordan canonical form is the following. ##### Theorem (Generalized eigenspace decomposition) Let $A$ be an $n\times n$ matrix over $\mathbb{C}$ with distinct eigenvalues $\lambda_1,\ldots,\lambda_q$. Then $\{F_{\lambda_1}, \ldots, F_{\lambda_q}\}$ is linearly independent, and $$ \mathbb{R}^n = F_{\lambda_1} \oplus \cdots \oplus F_{\lambda_q}. $$ That is to say, if we pick a basis $\beta_{\lambda_i}$ of $F_{\lambda_i}$ for $i = 1,\ldots, q$ and set $\beta = \beta_{\lambda_1} \cup \cdots \cup \beta_{\lambda_q}$, then $$ [f_A]_\beta^\beta = \begin{bmatrix} N_1 & O & \cdots & O \\ O & N_2 & \ddots & \vdots \\ \vdots & \ddots & \ddots & O \\ O & \cdots & O & N_q \end{bmatrix}. $$ Here we may set $f_{\lambda_i}$ as the resetriction $f_A\big|_{E_{\lambda_i}}$ such that $N_i = [f_{\lambda_i}]_{\beta_i}^{\beta_i}$. Note that each $N_i$ only has a single eigenvalue $\lambda_i$ whose algebraic multiplicity is the size of $N_1$. In other words, $N_i - \lambda_i I$ is a matrix with a single eigenvalue $0$. An $n\times n$ matrix $A$ such that $A^k = O$ for some $k$ is called a **nilpotent matrix**, whose eigenvalues are $0$ with multiplicity $n$. The minimum integer $d\geq 0$ such that $A^d = O$ is called the **index** of $A$. Therefore, $N_i - \lambda_i I$ is a nilpotent matrix, and we are going to focus on these matrices. ##### Theorem (Nilpotent matrix decomposition) Let $N$ be an $n\times n$ nilpotent matrix. Then there is a basis $\beta$ of $\mathbb{C}^n$ such that $[f_N]_\beta^\beta$ is a block diagonal matrix whose diagonal blocks are nilpotent Jordan blocks. Let $N - \lambda I$ be an $n\times n$ nilpotent matrix of index $d$. The steps for finding such a basis are as follows: 1. Find a basis $\beta_{d-1}$ for $\Col(A^{d-1})\cap \ker(A)$. 2. Expand $\beta_{d-1}$ to $\beta_{d-1}\cup\beta_{d-2}$ as a basis of $\Col(A^{d-2})\cap\ker(A)$. Keep doing this until we find a basis $\beta_{d-1}\cup\cdots\cup\beta_0$ of $\Col(A^0)\cap\ker(A) = \ker(A)$. 3. Start with $\beta = \emptyset$. For $k = d-1, \ldots, 0$ and each vector $\bv\in\beta_k$, solve $A^{k}\bx = \bv$ for $\bx$. Then add $\bv = A^{d-1}\bx, A^{d-2}\bx, \ldots, \bx$ to $\beta$. ## Side stories - read the minimal polynomial from the Jordan canonical form ## Experiments ##### Exercise 1 執行以下程式碼。 ```python ### code set_random_seed(0) print_ans = False b = 4 h = 3 ms = [choice(list(range(1, h + 1))) for i in range(b)] max_ms = max(ms) D = block_diagonal_matrix([jordan_block(0,ms[i]) for i in range(b)]) n = sum(ms) Q = random_good_matrix(n,n,n,2) A = Q * D * Q.inverse() print("n =", n) pretty_print(LatexExpr("A ="), A) for i in range(1, h + 1): pretty_print(LatexExpr(r"\operatorname{nul}(A - 0I)^{%s} ="%i), (A^i).nullity()) if print_ans: print("Jordan canonical form:") pretty_print(D) print("minimal polynomial =", x^max_ms) ``` ##### Exercise 1(a) 求出 $A$ 的喬丹標準型。 :::warning - [x] seed =78944556465 --> `seed = 78944556465` - [x] $nul$ --> $\nul$ ::: `seed = 78944556465`. $\begin{aligned} &A=\begin{bmatrix} -2 & 2 &-3 & 2 & 1 \\ -2 & 2 &-3 & 2 & 1 \\ 0 & 0 & 0 & 0 & 0 \\ 4 &-4 & 6 &-4 &-2 \\ -8 & 8 &-12 & 8 & 4 \end{bmatrix}, \\ &\nul(A-0I)^1=4,\\ &\nul(A-0I)^2=5,\\ &\nul(A-0I)^3=5.\\ \end{aligned}$ :::warning - [x] 標點 ::: Ans: 從 $\nul(A-0I)^1=4$ 和 $\nul(A-0I)^2=5$ 我們知 $0$ 重複 $5$ 次,有 $4$ 塊。因此 Jordan form $J_A= J_{0,2} \oplus J_{0,1} \oplus J_{0,1} \oplus J_{0,1}= \begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}$. ##### Exercise 1(b) 計算 $m_A(x)$。 :::warning - [x] jordan --> Jordan - [x] 等號放進數學模式 ::: Ans: 從矩陣 $A$ 的 Jordan form 我們可以看出 $m_A(x) = x^2$. ## Exercises ##### Exercise 2 令 $$ A = \begin{bmatrix} 3 & -17 & 23 & 1 & 5 \\ 1 & 18 & -21 & 0 & -5 \\ 5 & -3 & 5 & 3 & -1 \\ -10 & 69 & -90 & -6 & -17 \\ -19 & 56 & -69 & -13 & -7 \end{bmatrix}. $$ 已知 $A$ 的相異特徵值為 $2, 3$。 求出廣義特徵空間 $F_2$ 的一組基底 $\beta_2$、 及 $F_3$ 的一組基底 $\beta_3$、 並令 $\beta = \beta_2\cup\beta_3$。 求 $[f_A]_\beta^\beta$。 :::warning - [x] $dim$ --> $\dim$, $ker$ --> $\ker$, $span$ --> $\vspan$ - [x] 代入 $\lambda = 2$ 這句不用放到 `aligned` 裡 ::: :::info 這題沒有叫你完成 Jordan form。只叫你找廣義特徵空間的基底而已。 ::: Ans: 由矩陣計算機知 $\lambda = 2 ,3$ 代入 $\lambda = 2$ $\begin{aligned} &\dim(\ker(A-2I)) =\dim(\ker(\begin{bmatrix} 1 &-17 & 23 & 1 & 5 \\ 1 & 16 &-21 & 0 &-5 \\ 5 &-3 & 3 & 3 &-1 \\ -10 & 69 &-90 &-8 &-17 \\ -19 & 56 &-69 &-13 &-9 \end{bmatrix})\\ =&\dim(\vspan\{ \begin{bmatrix} 1 \\ \frac{-1}{4} \\ 0 \\ \frac{-1}{4} \\ 1 \end{bmatrix} \} ) =1 = \space 含\space 2\space 的有幾塊。 \end{aligned}$ $\begin{aligned} &\dim(\ker(A-2I)^2) =\dim(\ker(\begin{bmatrix} -6 &-9 & 14 &-3 & 5 \\ 7 & 22 &-31 & 3 & 9 \\ 6 & 9 &-14 & 3 &-5 \\ 12 & 40 &-56 & 5 &-16 \\ -7 & 25 &-29 &-5 &-4 \end{bmatrix})\\ =&\dim(\vspan\{ \begin{bmatrix} -\frac{1}{4} \\ 1 \\ \frac{3}{4} \\ 1 \\ 0 \end{bmatrix} ,\begin{bmatrix} 0 \\ -1 \\ -1 \\ 0 \\ 1 \end{bmatrix} \}) =2= 2 \space 的重複次數\\ &(A-2I) \begin{bmatrix} -\frac{1}{4} \\ 1 \\ \frac{3}{4} \\ 1 \\ 0 \end{bmatrix} =\begin{bmatrix} 1 \\ 0 \\ 1 \\ -4 \\ -4 \end{bmatrix} \\ &\beta_2=\{ \begin{bmatrix} -\frac{1}{4} \\ 1 \\ \frac{3}{4} \\ 1 \\ 0 \end{bmatrix} ,\begin{bmatrix} 1 \\ 0 \\ 1 \\ -4 \\ -4 \end{bmatrix} \} =\{ \bv_2 , \bv_1 \} \end{aligned}$ 代入 $\lambda = 3$ $\begin{aligned} &\dim(\ker(A-3I)) =\dim(\ker(\begin{bmatrix} 0 &-17 & 23 & 1 & 5 \\ 1 & 15 &-21 & 0 &-5 \\ 5 &-3 & 2 & 3 &-1 \\ -10 & 69 &-90 &-9 &-17 \\ -19 & 56 &-69 &-13 &-10 \end{bmatrix})\\ =&\dim(\vspan\{ \begin{bmatrix} 0 \\ \frac{1}{3} \\ 0 \\ \frac{2}{3} \\ 1 \end{bmatrix} \} ) =1 = \space 含\space 3\space 的有幾塊。 \end{aligned}$ $\begin{aligned} &\dim(\ker(A-3I)^3) =\dim(\ker(\begin{bmatrix} 10 &-35 & 45 & 7 & 7 \\ -3 & 10 &-13 &-2 &-2 \\ 7 &-25 & 32 & 5 & 5 \\ -40 & 140 &-180 &-28 &-28 \\ -37 & 130 &-167 &-26 &-26 \\ \end{bmatrix}))\\ =&\dim(\vspan\{ \begin{bmatrix} -1 \\ 1 \\ 1 \\ 0 \\ 0 \end{bmatrix},\begin{bmatrix} 0 \\ \frac{1}{5} \\ 0 \\ 1 \\ 0 \end{bmatrix},\begin{bmatrix} 0 \\ \frac{1}{5} \\ 0 \\ 0 \\ 1 \end{bmatrix}\})=3=\space 3 \space 的重複次數。\\ &(A-3I)\begin{bmatrix} -1 \\ 1 \\ 1 \\ 0 \\ 0 \end{bmatrix}=\begin{bmatrix} 6 \\ -7 \\ -6 \\ -11 \\ 6 \end{bmatrix},\\ &(A-3I)\begin{bmatrix} 6 \\ -7 \\ -6 \\ -11 \\ 6 \end{bmatrix}=\begin{bmatrix} 0 \\ -3 \\ 0 \\ -6 \\ -9 \end{bmatrix},\\ &\beta_3=\{ \begin{bmatrix} -1 \\ 1 \\ 1 \\ 0 \\ 0 \end{bmatrix},\begin{bmatrix} 6 \\ -7 \\ -6 \\ -11 \\ 6 \end{bmatrix},\begin{bmatrix} 0 \\ -3 \\ 0 \\ -6 \\ -9 \end{bmatrix} \}=\{\bv_5,\bv_4,\bv_3\}. \end{aligned}$ 因此 $\beta = \beta_2 \cup \beta_3=\{ \bv_1,\bv_2,\bv_3 ,\bv_4,\bv_5\}$, $[f_A]_\beta^\beta=J_{2,2} \oplus J_{3,3}=\begin{bmatrix} 2 &1 &0 &0 &0 \\ 0 &2 &0 &0 &0 \\ 0 &0 &3 &1 &0 \\ 0 &0 &0 &3 &1 \\ 0 &0 &0 &0 &3 \end {bmatrix}.$ ##### Exercise 3 令 $$ A = \begin{bmatrix} -2 & -9 & -1 & 5 & -5 \\ 4 & 20 & 1 & -9 & 10 \\ -3 & -9 & -2 & 7 & -6 \\ -10 & -47 & -4 & 24 & -25 \\ -16 & -78 & -5 & 37 & -40 \end{bmatrix}. $$ 已知 $A$ 為一冪零矩陣。 求一基底 $\beta$ 使得 $[f_A]_\beta^\beta$ 是喬丹標準型。 **[由鍾秉哲同學提供]** **1.** 已知 $A$ 為冪零矩陣,則其特徵值只有 $\lambda = 0$ 。 且經計算, $A ^ 3 = O$ , 推測 $A$ 的喬丹標準型為 **:** $$ \begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix} or \begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}. $$ 另外,透過高斯消去法, $A$ 有階梯形式 **:** $$ \begin{bmatrix} -2 & -9 & -1 & 5 & -5 \\ 0 & 2 & -1 & 1 & 0 \\ 0 & 0 & 7 & -11 & 6 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}. $$ 可知 $\operatorname{null}(A-0I) = 2$ , 因此 $A$ 的喬丹標準型為 **:** $$\begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}. $$ **2.** 設 $\bv_i$ 為 $A$ 的第 $i$ 個行向量。 觀察 **:** $$A^2 = \begin{bmatrix} -2 & -9 & -1 & 5 & -5 \\ 4 & 20 & 1 & -9 & 10 \\ -3 & -9 & -2 & 7 & -6 \\ -10 & -47 & -4 & 24 & -25 \\ -16 & -78 & -5 & 37 & -40 \end{bmatrix} \begin{bmatrix} -2 & -9 & -1 & 5 & -5 \\ 4 & 20 & 1 & -9 & 10 \\ -3 & -9 & -2 & 7 & -6 \\ -10 & -47 & -4 & 24 & -25 \\ -16 & -78 & -5 & 37 & -40 \end{bmatrix} = \begin{bmatrix} 1 & 2 & 0 & -1 & 1 \\ -1 & -2 & 0 & 1 & -1 \\ 2 & 4 & 0 & -2 & 2 \\ 4 & 8 & 0 & -4 & 4 \\ 5 & 10 & 0 & -5 & 5 \end{bmatrix} . $$ 可發現 $A \bv_3 = \bzero$ , 且 $A \bv_1 = -\bv_3$ , 也就是說, $$A ^ 3 \begin{bmatrix} -1 \\ 0 \\ 0 \\ 0 \\ 0 \end{bmatrix} = A^2 (-\bv_1) = A \bv_3 = \bzero . $$ 另外, $A \bv_1 = A \bv_5 = -\bv_3$ , 所以, $$ A (\bv_1 - \bv_5) = A \begin{bmatrix} 3 \\ -6 \\ 3 \\ 15 \\ 24 \end{bmatrix} = \bzero . $$ 也就是說, $$ A^2 \begin{bmatrix} 1 \\ 0 \\ 0 \\ 0 \\ -1 \end{bmatrix} = A \begin{bmatrix} 3 \\ -6 \\ 3 \\ 15 \\ 24 \end{bmatrix} = \bzero . $$ 且可以確認 $(\begin{bmatrix} -1 \\ 1 \\ -2 \\ -4 \\ -5 \end{bmatrix} , \begin{bmatrix} 3 \\ -6 \\ 3 \\ 15 \\ 24 \end{bmatrix})$ 線性獨立。 因此可取 $$ Q = \begin{bmatrix} -1 & 2& -1 & 3 & 1 \\ 1 & -4 & 0 & -6 & 0 \\ -2 & 3 & 0 & 3 & 0 \\ -4 & 10 & 0 & 15 & 0 \\ -5 & 16& 0 & 24 & -1 \end{bmatrix}. $$ 使得 $$ Q^{-1} A Q = \begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}. $$ ##### Exercise 4 凱力–漢米頓定理是證明喬丹標準型理論的關鍵之一。 然而假設已經知道每個複數矩陣都有喬丹標準型, 說明凱力–漢米頓定理在喬丹標準型的前提下是直觀的。 :::warning 懸賞 ::: ##### Exercise 5 令 $A$ 為一複數方陣。 說明 $e^A$ 會收斂。 :::warning 懸賞 ::: ##### Exercise 6 喬丹標準型的重要功用之一是判斷兩個矩陣是否相似。 在喬丹標準型的理論下不難發現以下敘述等價: 1. $A$ 和 $B$ 相似。 2. $A$ 和 $B$ 有相同的喬丹標準型。 判斷以下矩陣是否相似。 ##### Exercise 6(a) $$ A = \begin{bmatrix} 2 & 0 \\ -1 & 3 \end{bmatrix} \text{ and } B = \begin{bmatrix} 1 & 1 \\ -2 & 4 \end{bmatrix}. $$ **[由鍾秉哲同學提供]** $p_A (\lambda) = \det(A - \lambda I) = (2 - \lambda)(3 - \lambda)$ , 所以 $A$ 的喬丹標準型為 $$ \begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}. $$ $p_B (\lambda) = \det(B - \lambda I) = (1 - \lambda)(4 - \lambda) + 2 = - \lambda ^ 2 - 5 \lambda + 6 = (2 - \lambda)(3 - \lambda)$ , 所以 $B$ 的喬丹標準型為 $$ \begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}. $$ 因此 $A$ 與 $B$ 相似。 ##### Exercise 6(b) $$ A = \begin{bmatrix} 6 & -2 & -2 \\ 0 & 3 & 0 \\ 6 & -4 & -1 \end{bmatrix} \text{ and } B = \begin{bmatrix} 4 & -2 & 0 \\ 3 & 3 & 1 \\ -8 & 4 & 1 \end{bmatrix}. $$ **[由鍾秉哲同學提供]** $p_A (\lambda) = \det(A - \lambda I) = - \lambda ^ 3 + 8 \lambda ^ 2 - 21 \lambda +18 = (2 - \lambda) (3 - \lambda) ^ 2$ , 且 $\nul (A - 3I) = \nul (\begin{bmatrix} 3 & -2 & -2 \\ 0 & 0 & 0 \\ 6 & -4 & -4 \end{bmatrix}) = 2$ , 所以 $A$ 的喬丹標準型為 $$ \begin{bmatrix} 2 & 0 & 0 \\ 0 & 3 & 0 \\ 0 & 0 & 3 \end{bmatrix}. $$ $p_B (\lambda) = \det(B - \lambda I) = - \lambda ^ 3 + 8 \lambda ^ 2 - 21 \lambda +18 = (2 - \lambda) (3 - \lambda) ^ 2$ , 且 $\nul (B - 3I) = \nul (\begin{bmatrix} 1 & -2 & 0 \\ 3 & 0 & 1 \\ -8 & 4 & -2 \end{bmatrix}) = 1$ , 同時 $\nul ((B - 3I) ^ 2) = \nul (\begin{bmatrix} -5 & -2 & -2 \\ -5 & -2 & -2 \\ 20 & 8 & 8 \end{bmatrix}) = 2$ , 所以 $B$ 的喬丹標準型為 $$ \begin{bmatrix} 2 & 0 & 0 \\ 0 & 3 & 1 \\ 0 & 0 & 3 \end{bmatrix}. $$ 因此 $A$ 與 $B$ 不相似。 ##### Exercise 7 令 $A$ 為一 $n\times n$ 矩陣 ,$\lambda$ 為其一特徵值。 令 $F_\lambda$ 為其廣義特徵空間。 證明以下關於廣義特徵空間的性質。 ##### Exercise 7(a) 證明 $F_\lambda$ 是一個 $A$-不變子空間。 提示:$A(A - \lambda I)^k = (A - \lambda I)^k A$。 Ans: 設 ${\bf x} \in F_\lambda$,則可知對於某個 $k$ 有 $(A-\lambda I)^k {\bf x}={\bf 0}$。 考慮 ${\bf y}=A{\bf x}$,則有 $(A-\lambda I)^k {\bf y}=(A-\lambda I)^k A{\bf x}=A(A-\lambda I)^k {\bf x}=A{\bf 0}={\bf 0}$。 因此 ${\bf y}$ 也在 $F_\lambda$ 之中,即$F_\lambda$ 是一個 $A$-不變子空間。 :::success Great! ::: ##### Exercise 7(b) 令 $d$ 為 $(x - \lambda)$ 在 $m_A(x)$ 中的重數。 說明 $F_\lambda = \ker (A - \lambda I)^d$。 Ans: 已知 $d$ 為 $(x - \lambda)$ 在 $m_A(x)$ 中的重數,因此可知 $A$ 中最大的 $J_\lambda$ 是 $d$ 階。 對於 ${\bf x} \in F_\lambda$,都有 $k \leq d$ 使得 $(A-\lambda I)^k {\bf x}={\bf 0}$,即有 ${\bf x} \in \ker (A - \lambda I)^d$。 對於 ${\bf y} \in \ker (A - \lambda I)^d$,有 $\ker (A - \lambda I)^d {\bf y} = {\bf 0}$,因此 ${\bf y} \in F_\lambda$。 綜合以上可知 $F_\lambda = \ker (A - \lambda I)^d$。 :::info 這個證明在已知 Jordan form 和最小多項式的關係時是沒問題的, 但如果證明 Jordan form 需要用到這個性質,就會循環論證。 一般來說這題可以取任一個 $\bv \in F_\lambda$,然後要去看 $m_{A,\bv}(x)$。 因為 $m_{A,\bv}(x) \mid m_A(x)$,所以可以得到想要的結果。 ::: ##### Exercise 7(c) 令 $\alpha$ 為 $F_\lambda$ 的一組基底, 將其擴展為 $\mathbb{R}^n$ 的一組基底 $\beta$。 說明 $[f_A]_\beta^\beta$ 有以下型式 $$ [f_A]_\beta^\beta = \begin{bmatrix} N_\lambda & B \\ O & D \end{bmatrix}, $$ 其中 $N_\lambda$ 的特徵值均為 $\lambda$。 藉此說明 $\dim(F_\lambda) \leq \am(\lambda)$。 Ans: 設 $\alpha=\{ {\bf v}_1, \ldots , {\bf v}_k \}$,$\beta=\{ {\bf v}_1, \ldots , {\bf v}_n \}$。則 $$[f]_\beta^\beta = \begin{bmatrix} | & ~ & | & | & ~ & | \\ [f({\bf v}_1)]_\beta & \cdots & [f({\bf v}_k)]_\beta & [f({\bf v}_{k+1})]_\beta & \cdots & [f({\bf v}_n)]_\beta \\ | & ~ & | & | & ~ & | \\ \end{bmatrix} $$ 由前面的小題可知 $F_\lambda$ 是一個 $A$-不變子空間。對於 $1 \leq i \leq k$,有 ${\bf v}_i \in F_\lambda$。因此有$f({\bf v}_i)=A{\bf v}_i \in F_\lambda$,即 $f({\bf v}_i)$ 可用 $\alpha$ 內的向量的線性組合寫成,則 ${\bf v}_{k+1},\ldots,{\bf v}_n$ 的係數都為零,因此 $[f_A]_\beta^\beta$ 有以下型式 $$ [f_A]_\beta^\beta = \begin{bmatrix} N_\lambda & B \\ O & D \end{bmatrix}。 $$ ##### Exercise 7(d) 令 $A$ 的相異特徵值為 $\lambda_1, \ldots, \lambda_q$, 而 $F_{\lambda_1}, \ldots, F_{\lambda_q}$ 為它們的廣義特徵空間。 證明 $\{F_{\lambda_1}, \ldots, F_{\lambda_q}\}$ 線性獨立。 提示:令 $p_1(x)$ 為 $p_A(x)$ 中去掉所有 $(x - \lambda_1)$ 因式的多項式。 假設 $\bv_1 + \cdots + \bv_q = \bzero$ 且對於 $i = 1, \ldots, q$ 滿足 $\bv_i \in F_{\lambda_i}$。 將等號兩邊同乘 $p_1(A)$ 並利用 $m_{A,\bv}(x)$ 的性質來說明 $\bv_1 = \bzero$。 :::warning 懸賞 ::: ##### Exercise 7(e) 令 $$ p_A(x) = (\lambda_1 - x)^{\am(\lambda_1)} \cdots (\lambda_q - x)^{\am(\lambda_q)} $$ 且對每個 $i = 1,\ldots, q$ 計算 $B_i = (\lambda_i I - A)^{\am(\lambda_i)}$。 證明 $$ \nul(B_1) + \cdots + \nul(B_q) \geq n $$ 且 $$ \mathbb{R}^n = F_{\lambda_1} \oplus \cdots \oplus F_{\lambda_q}. $$ :::warning 懸賞 ::: ##### Exercise 8 令 $A$ 為一冪零矩陣且其深度(index)為 $d$。 證明以下關於冪零矩陣的性質。 ##### Exercise 8(a) 已知 $\{\bv_1,\bv_2,\bv_3\}$ 為 $\ker(A)$ 中的線性獨立集。 若存在 $\bx_1$ 及 $\bx_2$ 使得 $A\bx_1 = \bv_1$ 且 $A\bx_2 = \bv_2$。 說明 $\{\bv_1,\bv_2,\bv_3,\bx_1,\bx_2\}$ 線性獨立。 **$Ans:$** 利用否逆命題, 若 $\{\bv_1,\bv_2,\bv_3,\bx_1,\bx_2\}$ 不線性獨立, 則存在不全為 $0$ 的 $c_1$ ~ $c_5$ , 使 $c_1\bv_1 + c_2\bv_2 + c_3\bv_3 + c_4\bx_1 + c_5\bx_2 = \bzero$ , 且已知 $\{\bv_1,\bv_2,\bv_3\}$ 線性獨立, 則 $c_4 , c_5$ 至少有一不為 $0$ 。 將其代入 $f_A$ , $c_1A\bv_1 + c_2A\bv_2 + c_3A\bv_3 + c_4A\bx_1 + c_5A\bx_2 = A\bzero$ , 則 $c_4\bv_1 + c_5\bv_2 = \bzero$ , 也就是說, $\{\bv_1 , \bv_2\}$ 不線性獨立。 因此,若 $\{\bv_1,\bv_2,\bv_3\}$ 線性獨立, 可知 $\{\bv_1,\bv_2,\bv_3,\bx_1,\bx_2\}$ 也線性獨立。 :::success Good! ::: ##### Exercise 8(b) 定義函數一系列函數 $$ \mathbb{R}^n \xrightarrow{f_1} \Col(A) \xrightarrow{f_2} \Col(A^2) \xrightarrow{f_3}\cdots \xrightarrow{f_d} \Col(A^d) = \{\bzero\} $$ 其中對任意 $k = 1,\ldots, d$ 都定義 $f_k(\bx) = A\bx$。 證明對任意 $k = 1,\ldots, d$ 都有 $$ \rank(A^{k - 1}) = \rank(A^k) + \dim(\Col(A^{k-1})\cap\ker(A)), $$ 也就是 $$ \dim(\Col(A^{k-1})\cap\ker(A)) = \rank(A^{k - 1}) - \rank(A^k) = \nul(A^k) - \nul(A^{k - 1}). $$ :::info 這題的重點在於對 $f_k$ 使用維度定理,應該會簡單很多。 ::: **$Ans:$** 觀察 $\nul(A^k) - \nul(A^{k - 1}) = \dim(\ker(A^k) \cap \ker(A^{k - 1})^\perp )$ 。 由於任意 $\bu \in \ker(A^k) \cap \ker(A^{k - 1})^\perp$ 皆有 $A(A^{k-1} \bu) = A^{k} \bu = \bzero$ , 因此可令一個函數 $g$ : $$ \ker(A^k) \cap \ker(A^{k - 1})^\perp \rightarrow \Col(A^{k-1})\cap\ker(A) \\ \bu \mapsto A^{k-1} \bu $$ 其中,由於不存在 $\bu \neq \bzero$ 使 $g (\bu) = \bzero$ , 所以 $g$ 為 **injective** 。 同時,任意 $\bv \in \Col(A^{k-1})\cap\ker(A)$ 皆存在 $\bx$ 使 $A^{k-1} \bx = \bv$ 且 $A^{k} \bx =A (A^{k-1} \bx) = A \bv = \bzero$ , 所以 $g$ 為 **surjective** 。 可知 $g$ 可逆, 因此 $\dim (\ker(A^k) \cap \ker(A^{k - 1})^\perp ) = \dim (\Col(A^{k-1}) \cap \ker(A))$ , 得證 $\dim (\Col(A^{k-1}) \cap \ker(A)) = \dim (\ker(A^k) \cap \ker(A^{k - 1})^\perp ) = \nul(A^k) - \nul(A^{k - 1})$ 。 ##### Exercise 8(c) 證明 $$ \rank(A^{k - 1}) - \rank(A^k) = \nul(A^k) - \nul(A^{k - 1}) $$ 會隨著 $k$ 變大而遞減(或不變)。 :::info 證明應該是對,不過 $\Col(A^{k+1}) \subseteq \Col(A^k)$ 應該是比較直接的手法。 ::: **$Ans:$** 已知 $\nul(A^k) - \nul(A^{k - 1}) = \dim (\Col(A^{k-1}) \cap \ker(A))$ , 則需證 : 任意整數 $h \geqslant 2$ 皆有 $\nul(A^h) - \nul(A^{h - 1}) = \dim (\Col(A^{h-1}) \cap \ker(A)) \geqslant \dim (\Col(A^{h}) \cap \ker(A)) = \nul(A^{h + 1}) - \nul(A^{h})$ 。 設 $\beta = \{\bv_1 , \bv_2 , \ldots , \bv_n\}$ 為 $\Col(A^{k-1})$ 的基底, 由於任意 $\bu \in \Col(A^k)$ 皆有 $\bv \in \Col(A^{k - 1})$ 使 $A \bv = \bu$ , 因此,存在 $c_1 , c_2 , \ldots , c_n \in \mathbb{R}$ 使 $\bu = A \bv =A (c_1 \bv_1 + c_2 \bv_2 + \ldots + c_n \bv_n) = c_1 A\bv_1 + c_2 A\bv_2 + \ldots + c_n A\bv_n$ , 也就是說,任意 $\bu \in \Col(A^k)$ 皆能以 $\{A \bv_1 , A \bv_2 , \ldots , A \bv_n\}$ 表示, 所以 $\dim ( \Col(A^{h}) ) \leqslant n = \dim (\Col(A^{h-1}))$ , 得證 $\dim (\Col(A^{h-1}) \cap \ker(A)) \geqslant \dim (\Col(A^{h}) \cap \ker(A))$ 。 :::info 分數 = 6.5 :::

    Import from clipboard

    Paste your markdown or webpage here...

    Advanced permission required

    Your current role can only read. Ask the system administrator to acquire write and comment permission.

    This team is disabled

    Sorry, this team is disabled. You can't edit this note.

    This note is locked

    Sorry, only owner can edit this note.

    Reach the limit

    Sorry, you've reached the max length this note can be.
    Please reduce the content or divide it to more notes, thank you!

    Import from Gist

    Import from Snippet

    or

    Export to Snippet

    Are you sure?

    Do you really want to delete this note?
    All users will lose their connection.

    Create a note from template

    Create a note from template

    Oops...
    This template has been removed or transferred.
    Upgrade
    All
    • All
    • Team
    No template.

    Create a template

    Upgrade

    Delete template

    Do you really want to delete this template?
    Turn this template into a regular note and keep its content, versions, and comments.

    This page need refresh

    You have an incompatible client version.
    Refresh to update.
    New version available!
    See releases notes here
    Refresh to enjoy new features.
    Your user state has changed.
    Refresh to load new user state.

    Sign in

    Forgot password

    or

    By clicking below, you agree to our terms of service.

    Sign in via Facebook Sign in via Twitter Sign in via GitHub Sign in via Dropbox Sign in with Wallet
    Wallet ( )
    Connect another wallet

    New to HackMD? Sign up

    Help

    • English
    • 中文
    • Français
    • Deutsch
    • 日本語
    • Español
    • Català
    • Ελληνικά
    • Português
    • italiano
    • Türkçe
    • Русский
    • Nederlands
    • hrvatski jezik
    • język polski
    • Українська
    • हिन्दी
    • svenska
    • Esperanto
    • dansk

    Documents

    Help & Tutorial

    How to use Book mode

    Slide Example

    API Docs

    Edit in VSCode

    Install browser extension

    Contacts

    Feedback

    Discord

    Send us email

    Resources

    Releases

    Pricing

    Blog

    Policy

    Terms

    Privacy

    Cheatsheet

    Syntax Example Reference
    # Header Header 基本排版
    - Unordered List
    • Unordered List
    1. Ordered List
    1. Ordered List
    - [ ] Todo List
    • Todo List
    > Blockquote
    Blockquote
    **Bold font** Bold font
    *Italics font* Italics font
    ~~Strikethrough~~ Strikethrough
    19^th^ 19th
    H~2~O H2O
    ++Inserted text++ Inserted text
    ==Marked text== Marked text
    [link text](https:// "title") Link
    ![image alt](https:// "title") Image
    `Code` Code 在筆記中貼入程式碼
    ```javascript
    var i = 0;
    ```
    var i = 0;
    :smile: :smile: Emoji list
    {%youtube youtube_id %} Externals
    $L^aT_eX$ LaTeX
    :::info
    This is a alert area.
    :::

    This is a alert area.

    Versions and GitHub Sync
    Get Full History Access

    • Edit version name
    • Delete

    revision author avatar     named on  

    More Less

    Note content is identical to the latest version.
    Compare
      Choose a version
      No search result
      Version not found
    Sign in to link this note to GitHub
    Learn more
    This note is not linked with GitHub
     

    Feedback

    Submission failed, please try again

    Thanks for your support.

    On a scale of 0-10, how likely is it that you would recommend HackMD to your friends, family or business associates?

    Please give us some advice and help us improve HackMD.

     

    Thanks for your feedback

    Remove version name

    Do you want to remove this version name and description?

    Transfer ownership

    Transfer to
      Warning: is a public team. If you transfer note to this team, everyone on the web can find and read this note.

        Link with GitHub

        Please authorize HackMD on GitHub
        • Please sign in to GitHub and install the HackMD app on your GitHub repo.
        • HackMD links with GitHub through a GitHub App. You can choose which repo to install our App.
        Learn more  Sign in to GitHub

        Push the note to GitHub Push to GitHub Pull a file from GitHub

          Authorize again
         

        Choose which file to push to

        Select repo
        Refresh Authorize more repos
        Select branch
        Select file
        Select branch
        Choose version(s) to push
        • Save a new version and push
        • Choose from existing versions
        Include title and tags
        Available push count

        Pull from GitHub

         
        File from GitHub
        File from HackMD

        GitHub Link Settings

        File linked

        Linked by
        File path
        Last synced branch
        Available push count

        Danger Zone

        Unlink
        You will no longer receive notification when GitHub file changes after unlink.

        Syncing

        Push failed

        Push successfully