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
    • 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 Versions and GitHub Sync Note Insights 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
    Subscribed
    • Any changes
      Be notified of any changes
    • Mention me
      Be notified of mention me
    • Unsubscribe
    Subscribe
    {%hackmd 5xqeIJ7VRCGBfLtfMi0_IQ %} # 代數重數與幾何重數 ![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_int_list, random_good_matrix ``` ## Main idea We have seen whether a matrix $A$ is diagonalizable depends on how many (independent) vectors we can find in $\ker(A - \lambda I)$ for each eigenvalue $\lambda\in\spec(A)$. In this section, we will give some quantitative measures on how much a matrix is diagonalizable. Let $A$ be an $n\times n$ matrix and $p_A(x)$ its characteristic polynomial. For each $\lambda \in \spec(A)$, - the **algebtraic multiplicity** of $\lambda$ is the multiplicity of $\lambda$ as a root of $p_A(x)$, denoted by $\am_A(\lambda)$, while - the **geometric multiplicity** of $\lambda$ is the dimension of $\ker(A - \lambda I)$, denoted by $\gm_A(\lambda)$. When the context is clear, the subscript $A$ can be omitted. For any $n\times n$ matrix $A$ and $\lambda\in\spec(A)$, the following properties hold. - $1 \leq \gm(\lambda) \leq \am(\lambda)$. - The sum of $\am(\lambda)$ over all distinct eigenvalues $\lambda$ is $n$. - The matrix $A$ is diagonalizable if and only $\gm(\lambda) = \am(\lambda)$ for all distinct eigenvalues $\lambda$. The fact $1\leq\gm(\lambda)$ follows immediate from the definition that $A - \lambda I$ is singular when $\lambda$ is an eigenvalue. The exercises contains a proof of $\gm(\lambda) \leq \am(\lambda)$. The sum of $\am(\lambda)$ is equal to the degree of $p_A(x)$, which is $n$. If $\gm(\lambda) < \am(\lambda)$ for some $\lambda$, then the sum of all geometric multiplicities is less than $n$, so it is impossible to find a basis composed of eigenvectors. If $\gm(\lambda) = \am(\lambda)$ for all distinct eigenvalues $\lambda$, we will show in the next section that $A$ is diagonalizable. ##### Proposition Let $A$ be an $n\times n$ matrix. If the characteristic polynomial $p_A(x)$ has $n$ distinct roots, then $A$ is diagonalizable. ## Side stories - Jordan block - invariant subspace ## Experiments ##### Exercise 1 執行以下程式碼。 ```python ### code set_random_seed(0) print_ans = False am1 = choice([2,3]) am2 = choice([2,3]) n = am1 + am2 while True: lam1, lam2 = random_int_list(2,3) if lam1 != lam2: break gm1 = randint(1,am1) gm2 = randint(1,am2) block1 = [matrix([[lam1]]) for i in range(gm1 - 1)] + [jordan_block(lam1, am1 - gm1 + 1)] block2 = [matrix([[lam2]]) for i in range(gm2 - 1)] + [jordan_block(lam2, am2 - gm2 + 1)] D = block_diagonal_matrix(block1 + block2) Q = random_good_matrix(n,n,n,2) A = Q * D * Q.inverse() pretty_print(LatexExpr("A ="), A) pA = (-1)^n * A.charpoly() print("characteristic polyomial =", pA) print(" =", factor(pA)) print("rref of A - (%s)I:"%lam1) pretty_print((A - lam1).rref()) print("rref of A - (%s)I:"%lam2) pretty_print((A - lam2).rref()) if print_ans: print("lambda =", lam1) print("am(lambda) =", am1) print("gm(lambda) =", gm1) print("lambda =", lam2) print("am(lambda) =", am2) print("gm(lambda) =", gm2) print("Diagonalizable?", am1 == gm1 and am2 == gm2) ``` ##### Exercise 1(a) 求 $A$ 的每一個特徵值的代數重數與幾何重數。 $Ans:$ 當 ```seed=0``` 時, $$ A = \begin{bmatrix} 19 & -23 & -3 & 9 \\ -38 & 40 & 7 & -17 \\ -77 & 85 & 11 & -34 \\ -173 & 190 & 29 & -78 \\ \end{bmatrix}. $$ $p_A(x)=x^4 + 8x^3 + 22x^2 + 24x + 9=(x + 1)^2 (x + 3)^2。$ 若 $p_A(x)=0$ , 則 $x=-1,-1,-3,-3$ , $\spec(A) = \{-1,-1,-3,-3\}$。 $\lambda = -1$ , $A-\lambda I = \begin{bmatrix} 20 & -23 & -3 & 9 \\ -38 & 41 & 7 & -17 \\ -77 & 85 & 12 & -34 \\ -173 & 190 & 29 & -77 \\ \end{bmatrix}$ , $(A-\lambda I)$ 的 RREF 為 $\begin{bmatrix} 1 & 0 & 0 & \frac{2}{17}\\ 0 & 1 & 0 & -\frac{4}{17}\\ 0 & 0 & 1 & -\frac{7}{17}\\ 0 & 0 & 0 & 0 \end{bmatrix}$, $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}-\frac{2}{17}\\\frac{4}{17}\\\frac{7}{17}\\1\end{pmatrix}\right\}$; $\lambda = -3$ , $A-\lambda I = \begin{bmatrix} 22 & -23 & -3 & 9 \\ -38 & 43 & 7 & -17 \\ -77 & 85 & 14 & -34 \\ -173 & 190 & 29 & -75 \\ \end{bmatrix}$ , $A-\lambda I$ 的 RREF 為 $\begin{bmatrix} 1 & 0 & 0 & \frac{1}{6}\\ 0 & 1 & 0 & -\frac{1}{6}\\ 0 & 0 & 1 & -\frac{1}{2}\\ 0 & 0 & 0 & 0\\ \end{bmatrix}$ $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}-\frac{1}{6}\\\frac{1}{6}\\\frac{1}{2}\\1\end{pmatrix}\right\}$ 。 一個特徵值的代數重數為該特徵值在特徵多項式中重根幾次。 一個特徵值 $\lambda$ 的幾何重數為 $\operatorname{null}(A-\lambda I)$。 則我們可以得到 $\am_A(-1) = 2$ , $\gm_A(-1) = 1$ , $\am_A(3) = 2$ , $\gm_A(3) = 1$ ##### Exercise 1(b) 判斷 $A$ 是否可對角化。 $Ans:$ 如果 $A$ 的所有特徵值的代數重數等於幾何重數,那麼 $A$ 可以對角化. 如果 $A$ 的其中一個特徵值的代數重數不等於幾何重數,那麼 $A$ 不可以對角化. 因為 $\am_A(-1) \not = \gm_A(-1)$ 且 $\am_A(-3) \not = \gm_A(-3)$ ,所以 $A$ 不可對角化。 ## Exercises ##### Exercise 2 對以下矩陣, 計算每一個特徵值的代數重數與幾何重數, 並判斷其是否可以對角化。 ##### Exercise 2(a) $$ A = \begin{bmatrix} 0 & 1 \\ -6 & 5 \end{bmatrix}. $$ $Ans:$ $$ \begin{aligned} p_A(x) &= \det(A-xI)\\ &= -x(5-x) + 6\\ &= x^2 - 5x + 6\\ &= (x-2)(x-3), \end{aligned} $$ 若 $p_A(x)=0$ , 則$x=3,2$ , $\spec(A) = \{3,2\}$。 $\lambda = 3$ , $A-\lambda I = \begin{bmatrix} -3 & 1 \\ -6 & 2 \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}\frac{1}{3}\\1\end{pmatrix}\right\}$ ; $\lambda = 2$ , $A-\lambda I = \begin{bmatrix} -2 & 1 \\ -6 & 3 \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}\frac{1}{2}\\1\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(3) = \gm_A(3) = 1, \am_A(2) = \gm_A(2)=1$ , 又對於所有相異的 $\lambda$ 都滿足 $\am_A(\lambda) = \gm_A(\lambda)$ ,所以 $A$ 可對角化。 ##### Exercise 2(b) $$ A = \begin{bmatrix} -1 & 1 \\ -1 & 1 \end{bmatrix}. $$ $Ans:$ $$ \begin{aligned} p_A(x) &= \det(A-xI)\\ &= (-1-x)(1-x)+1\\ &= x^2, \end{aligned} $$ 若$p_A(x)=0$ , $x=0,0$ , $\spec(A) = \{0,0\}$ 。 $\lambda = 0$ , $A-\lambda I = \begin{bmatrix} -1 & 1 \\ -1 & 1 \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}1\\1\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(0) = 2, \gm_A(0) = 1$ ,因為 $\am_A(0) \not = \gm_A(0)$ ,所以 $A$ 不可對角化。 ##### Exercise 3 對以下矩陣, 計算每一個特徵值的代數重數與幾何重數, 並判斷其是否可以對角化。 ##### Exercise 3(a) $$ A = \begin{bmatrix} 4 & 0 & -1 \\ 0 & 4 & -1 \\ -1 & -1 & 5 \end{bmatrix}. $$ $Ans:$ 找 $p_A(x) = \det(A-xI) = -(x-3)(x-4)(x-6)$ ,若 $p_A(x) = 0$ ,即可得 $\spec(A) = \{3,4,6\}$ 。 令 $\lambda = 3$ , $A-\lambda I = \begin{bmatrix} 1 & 0 & -1 \\ 0 & 1 & -1 \\ -1 & -1 & 2 \\ \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}1\\1\\1\end{pmatrix}\right\}$ ; $\lambda = 4$ , $A-\lambda I = \begin{bmatrix} 0 & 0 & -1 \\ 0 & 0 & -1 \\ -1 & -1 & 1 \\ \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}-1\\1\\0\end{pmatrix}\right\}$ ; $\lambda = 6$ , $A-\lambda I = \begin{bmatrix} -2 & 0 & -1 \\ 0 & -2 & -1 \\ -1 & -1 & -1 \\ \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}-\frac{1}{2}\\-\frac{1}{2}\\1\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(3) = \gm_A(3) = 1, \am_A(4) = \gm_A(4) = 1, \am_A(6) = \gm_A(6) = 1$ , 又對於所有相異的 $\lambda$ 都滿足 $\am_A(\lambda) = \gm_A(\lambda)$ ,所以 $A$ 可對角化。 ##### Exercise 3(b) $$ A = \begin{bmatrix} 1 & 3 & 2 \\ 2 & 4 & 3 \\ -3 & -7 & -5 \end{bmatrix}. $$ $Ans:$ 找 $p_A(x) = \det(A-xI) = -x^3$ ,若 $p_A(x) = 0$ ,即可得 $\spec(A) = \{0,0,0\}$ 。 令 $\lambda = 0$ , $A-\lambda I = \begin{bmatrix} 1 & 3 & 2 \\ 2 & 4 & 3 \\ -3 & -7 & -5 \\ \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}-\frac{1}{2}\\-\frac{1}{2}\\1\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(0) = 3, \gm_A(0) = 1$ ,因為 $\am_A(0) \not = \gm_A(0)$ ,所以 $A$ 不可對角化。 ##### Exercise 4 以下矩陣稱為喬丹區塊矩陣(Jordan block)。 說明以下矩陣 不計算重數的話只有一個特徵值, 計算這個特徵值的代數重數與幾何重數。 因此大小大於等於 $2$ 的喬丹區塊矩陣皆不可對角化。 ##### Exercise 4(a) $$ A = \begin{bmatrix} 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix}. $$ $Ans:$ 找 $p_A(x) = \det(A-xI) = x^4$ ,若 $p_A(x) = 0$ ,即可得 $\spec(A) = \{0,0,0,0\}$ 。 令 $\lambda = 0$ , $A-\lambda I = \begin{bmatrix} 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}1\\0\\0\\0\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(0) = 4, \gm_A(0) = 1$ ,因為 $\am_A(0) \not = \gm_A(0)$ ,所以 $A$ 不可對角化。 ##### Exercise 4(b) $$ A = \begin{bmatrix} 3 & 1 & 0 & 0 & 0 \\ 0 & 3 & 1 & 0 & 0 \\ 0 & 0 & 3 & 1 & 0 \\ 0 & 0 & 0 & 3 & 1 \\ 0 & 0 & 0 & 0 & 3 \end{bmatrix}. $$ $Ans:$ 找 $p_A(x) = \det(A-xI) = (3-x)^5$ ,若 $p_A(x) = 0$ ,即可得 $\spec(A) = \{3,3,3,3,3\}$ 。 令 $\lambda = 3$ , $A-\lambda I = \begin{bmatrix} 0 & 1 & 0 & 0 & 0 \\ 0 & 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix}$ , $\ker(A-\lambda I) = \vspan\left\{\begin{pmatrix}1\\0\\0\\0\\0\end{pmatrix}\right\}$ 。 則我們可以得到 $\am_A(3) = 5, \gm_A(3) = 1$ ,因為 $\am_A(3) \not = \gm_A(3)$ ,所以 $A$ 不可對角化。 ##### Exercise 5 找尋滿足以下條件的矩陣。 ##### Exercise 5(a) 找一個矩陣 $A$ 使得 $\spec(A) = \{2,2,3,3,3\}$、 $\gm(2) = 1$ 且 $\gm(3) = 2$。 **ANS:** $A = \begin{bmatrix} 2 & 0 & 0 & 0 & 0 \\ 1 & 3 & 0 & 0 & 0 \\ 0 & 1 & 2 & 0 & 0 \\ 0 & 0 & 1 & 3 & 0 \\ 0 & 0 & 0 & 0 & 3 \end{bmatrix}$。 \ 再來透過驗算可以得到以下結果: \ 當 $\lambda=2$ 時, $\ker(A-2 I) = \vspan\left\{\begin{pmatrix}0\\0\\0\\-1\\1\end{pmatrix}\right\}$ 。 \ 因此 $\gm(2) = 1$。 \ 當 $\lambda=3$ 時,特徵向量為$\begin{bmatrix} 0 \\ 0 \\ 0 \\ 1 \\ 0 \end{bmatrix}$ 以及$\begin{bmatrix} 0 \\ 0 \\ 0 \\ 0 \\ 1 \end{bmatrix}$, 因此 $\gm(3) = 2$。 :::success Good job! 我心裡想的是 $J_{2,2}\oplus J_{3,2} \oplus J_{3,1}$。 ::: ##### Exercise 5(b) 找一個每一項都不為零的矩陣 $A$ 使得 $\spec(A) = \{2,2,3,3,3\}$、 $\gm(2) = 1$ 且 $\gm(3) = 2$。 **Ans:** 令 $$ A = \begin{bmatrix} 6 & -2 & 16 & -10 & -7\\ -2 & 1 & 6 & -3 & -2\\ 4 & -7 & 43 & -24 & -17\\ 5 & -10 & 60 & -33 & -25\\ 5 & -2 & 14 & -9 & -4 \end{bmatrix}. $$ 驗證: $$ \begin{aligned} p_A(x) &= \det(A-xI)\\ &=-x^5+13x^4-67x^3+171x^2-216x+108\\ &=-(x-2)^2(x-3)^3 \end{aligned} $$ 所以 $\spec(A) = \{ 2, 2, 3, 3, 3 \}$,$\am(2)=2$,$\am(3)=3$. $A-2I$ 的 RREF 為 $$ \begin{bmatrix} 1 & 0 & 0 & -\frac{1}{5} & 0\\ 0 & 1 & 0 & -\frac{1}{5} & 0\\ 0 & 0 & 1 & -\frac{3}{5} & 0\\ 0 & 0 & 0 & 0 & 1\\ 0 & 0 & 0 & 0 & 0 \end{bmatrix} $$ 所以 $\gm(2) = 1$. $A-3I$ 的 RREF 為 $$ \begin{bmatrix} 1 & 0 & 0 & -\frac{2}{15} & \frac{1}{3}\\ 0 & 1 & 0 & -\frac{4}{15} & \frac{1}{3}\\ 0 & 0 & 1 & -\frac{19}{30} & -\frac{1}{3}\\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \end{bmatrix} $$ 所以 $\gm(3) = 2$. :::success :+1: ::: ##### Exercise 6 令 $A$ 為一 $n\times n$ 矩陣而 $\lambda$ 為其一特徵值。 依照以下步驟證明 $\gm(\lambda) \leq \am(\lambda)$。 ##### Exercise 6(a) 令 $d = \gm(\lambda)$ 為 $\ker(A - \lambda I)$ 的維度。 找一組 $\ker(A - \lambda I)$ 的基底 $\alpha$, 並將其擴展為 $\mathbb{R}^n$ 的一組基底 $\beta$。 (所以 $\alpha\subseteq\beta$ 且 $|\alpha| = d$。) 說明 $[f_A]_\beta^\beta$ 會有以下型式 $$ \begin{bmatrix} \lambda I_d & B \\ O_{n-d,d} & D \end{bmatrix}. $$ (這裡 $D$ 只是一個矩陣的符號,不一定是對角矩陣。) :::warning - [x] $\lambda_i$ --> $\lambda$ - [x] 說明 $Q, Q_1, Q_2$ 是什麼 - [x] 都沒用到 $\alpha$ 耶?條件沒用到但是證出來,不是題目錯就是證明有錯 :astonished: - [ ] $k$ --> $d$ - [ ] 其中 $v_1,v_k$ 為 --> 其中 $\bv_1,\ldots, \bv_k$ 皆為 - [ ] 向量用粗體 $\bv$ - [ ] 取 $P=[ v_1, \cdots, v_k, v_{k+1}, \cdots, v_n ]=[UV]$ 為可逆矩陣,其中 $U=\alpha$, $V=\beta$。 --> 取 $P$ 為由 $\beta$ 元素當作行向量的矩陣、$U$ 為由 $\alpha$ 元素當作行向量的矩陣,則 $P$ 可寫為 $\begin{bmatrix} U & V \end{bmatrix}$. - [ ] $\begin{bmatrix} UV \end{bmatrix}$ --> $\begin{bmatrix} U & V \end{bmatrix}$ - [ ] 整串的數學式可以用一整個 `$...$` 包起來就好,不用進進出出數學模式,像是 $$ \begin{aligned} P^{-1}AP &= \begin{bmatrix} X \\ Y \end{bmatrix} A \begin{bmatrix} U & V \end{bmatrix} = \begin{bmatrix} XAU & XAV \\ YAU & YAV \end{bmatrix} \\ &= \begin{bmatrix} \lambda XU & XAV \\ \lambda YU & XAV \end{bmatrix} = \begin{bmatrix} \lambda I_d & XAV \\ O & YAV \end{bmatrix}. \end{aligned} $$ ::: **ANS1:** 令 $\alpha = \{ v_1, \cdots, v_k \}$,其中 $v_1,v_k$ 為 $A$ 相對於 $\lambda$ 的特徵向量。 則存在 $v_{k+1}, \cdots, v_n$ 使得 $\beta = \{ v_1, \cdots, v_k, v_{k+1}, \cdots, v_n \}$為 $\mathbb{R}^n$ 的一組基底。 取 $P=[ v_1, \cdots, v_k, v_{k+1}, \cdots, v_n ]=[UV]$ 為可逆矩陣,其中 $U=\alpha$, $V=\beta$。 \ 令 $P^{-1} = \begin{bmatrix} X \\ Y \end{bmatrix}$,其中 $X$ 為 $d\times n$ 矩陣,$Y$ 為 $(n-d) \times n$矩陣。 \ 則 $P^{-1}P=\begin{bmatrix} X \\ Y \end{bmatrix}$$\begin{bmatrix} UV \end{bmatrix}$$=\begin{bmatrix} XU & XV \\ YU & YV \end{bmatrix}=\begin{bmatrix} I_d & O \\ O & I_{n-k} \end{bmatrix}.$ \ 由上式得到 $XU=I_d,XV=O,YU=O,YV=I_{n-d}$。 \ 再細看可以得知$AU=A\{ v_1, \cdots, v_k \}=\{ Av_1, \cdots, Av_k \}=\{ \lambda v_1, \cdots, \lambda v_k \}=\lambda U$。 \ 因此,$P^{-1}AP=\begin{bmatrix} X \\ Y \end{bmatrix}A$$\begin{bmatrix} UV \end{bmatrix}$$=\begin{bmatrix} XAU & XAV \\ YAU & YAV \end{bmatrix}=\begin{bmatrix} \lambda XU & XAV \\ \lambda YU & XAV \end{bmatrix}$=\begin{bmatrix} \lambda I_d & XAV \\ O & YAV \end{bmatrix} \ 其中$B,D$為任意矩陣,故得證。 **ANS2:** 令 $\alpha = \{ {\bf \alpha}_1, {\bf \alpha}_2, \cdots, {\bf \alpha}_d \}$,$\beta = \{ {\bf \alpha}_1, {\bf \alpha}_2, \cdots, {\bf \alpha}_d, {\bf \beta}_1, {\bf \beta}_2, \cdots, {\bf \beta}_{n-d} \}$, $$ M = \begin{bmatrix} | & & | & | & & |\\ {\bf \alpha}_1 & \cdots & {\bf \alpha}_d & {\bf \beta}_1 & \cdots & {\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}. $$ 因為特徵向量 ${\bf \alpha}_i$, $i=1,2,\ldots,d$ 對應到特徵值 $\lambda$,且 $A{\bf \alpha}_i = \lambda {\bf \alpha}_i$, $$ AM = \begin{bmatrix} | & & | & | & & |\\ \lambda{\bf \alpha}_1 & \cdots & \lambda{\bf \alpha}_d & A{\bf \beta}_1 & \cdots & A{\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}. $$ 觀察 $M^{-1}M = I_n$,可以知道 $M^{-1}{\bf \alpha}_i = (0,\cdots,0,1,0,\cdots,0)\trans$ 是第 $i$ 個位置是 $1$,其它為 $0$ 的向量. 那麼只關心前 $d$ 行的話 $$ \begin{aligned} M^{-1}AM &= [f_A]_\beta^\beta\\ &=\begin{bmatrix} | & & | & | & & |\\ M^{-1}\lambda{\bf \alpha}_1 & \cdots & M^{-1}\lambda{\bf \alpha}_d & M^{-1}A{\bf \beta}_1 & \cdots & M^{-1}A{\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}\\ &=\begin{bmatrix} | & & | & | & & |\\ \lambda M^{-1}{\bf \alpha}_1 & \cdots & \lambda M^{-1}{\bf \alpha}_d & M^{-1}A{\bf \beta}_1 & \cdots & M^{-1}A{\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}\\ &=\begin{bmatrix} \lambda I_d & B \\ O_{n-d,d} & D \end{bmatrix}. \end{aligned} $$ 其中 $B,D$ 是不重要的兩個矩陣. :::success 寫得不錯,不過應該蠻好看出 $\lambda\alpha_i$ 的表示法就是一個全零的向量、並把第 $i$ 項改成 $\lambda$;當然用 $M^{-1}$ 也行。 ::: ##### Exercise 6(b) 由於 $[f_A]_\beta^\beta$ 和 $A$ 有相同的特徵多項式。 說明 $p_A(x)$ 中有 $(\lambda - x)^d$ 這個因式, 因此 $\gm(\lambda) \leq \am(\lambda)$。 **Ans:** 由於 $[f_A]_\beta^\beta$ 和 $A$ 有相同的特徵多項式,所以 $$ \begin{aligned} p_A(x) &= \det([f_A]_\beta^\beta - xI)\\ \\ &= \det( \begin{bmatrix} (\lambda - x)I_d & B\\ O_{n-d,d} & D' \end{bmatrix} )\\ \\ &= \det((\lambda-x)I_d) \det(D')\\ \\ &= (\lambda-x)^d \det(D'). \end{aligned} $$ 其中 $D'$ 是 $D$ 對角線上元素減 $x$. 可以知道 $\lambda$ 至少重根 $d$ 次,所以 $\am(\lambda) \geq d = \gm(\lambda)$. ##### Exercise 7 令 $A$ 為一矩陣而 $W$ 為一 $\mathbb{R}^n$ 的子空間。 如果 $$ f_A(W) = \{A\bw: \bw\in W\} \subseteq W, $$ 則我們稱 $W$ 是一個 **$A$-不變子空間($A$-invariant subspace)**。 同樣地,令 $f:V\rightarrow V$ 為一個線性函數 而 $W$ 是 $V$ 的一個子空間。 如果 $$ f(W) = \{f(\bw): \bw\in W\} \subseteq W, $$ 則我們稱 $W$ 是一個 **$f$-不變子空間($f$-invariant subspace)**。 ##### Exercise 7(a) 令 $A$ 為一 $n\times n$ 矩陣而 $W$ 為一 $A$-不變子空間。 令 $\alpha$ 為 $W$ 的一組基底 並將基擴展為 $\mathbb{R}^n$ 的一組基底 $\beta$。 說明 $[f_A]_\beta^\beta$ 會有以下型式 $$ \begin{bmatrix} C & B \\ O_{n-d,d} & D \end{bmatrix}. $$ (這裡 $D$ 只是一個矩陣的符號,不一定是對角矩陣。) **Ans:** 令 $\alpha = \{ {\bf \alpha}_1, {\bf \alpha}_2, \cdots, {\bf \alpha}_d \}$,$\beta = \{ {\bf \alpha}_1, {\bf \alpha}_2, \cdots, {\bf \alpha}_d, {\bf \beta}_1, {\bf \beta}_2, \cdots, {\bf \beta}_{n-d} \}$, $$ M = \begin{bmatrix} | & & | & | & & |\\ {\bf \alpha}_1 & \cdots & {\bf \alpha}_d & {\bf \beta}_1 & \cdots & {\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}. $$ 因為 $W$ 為一 $A$-不變子空間,所以 $A {\bf \alpha}_i \in W$,$i = 1,2,\ldots,d$,並且有唯一的表達式,$A {\bf \alpha}_i = c_{i,1}{\bf \alpha}_1 + c_{i,2}{\bf \alpha}_2 + \cdots + c_{i,d}{\bf \alpha}_d$. $$ \begin{aligned} M^{-1}A{\bf \alpha}_i &= c_{i,1}M^{-1}{\bf \alpha}_1 + c_{i,2}M^{-1}{\bf \alpha}_2 + \cdots + c_{i,d}M^{-1}{\bf \alpha}_d\\ \\ &= \begin{pmatrix} c_{i,1}\\c_{i,2}\\ \vdots\\ c_{i,d}\\0\\0\\ \vdots\\ 0 \end{pmatrix}. \end{aligned} $$ 如果只關心前 $d$ 行的話, $$ \begin{aligned} M^{-1}AM &= [f_A]_\beta^\beta\\ &=\begin{bmatrix} | & & | & | & & |\\ M^{-1}A{\bf \alpha}_1 & \cdots & M^{-1}A{\bf \alpha}_d & M^{-1}A{\bf \beta}_1 & \cdots & M^{-1}A{\bf \beta}_{n-d}\\ | & & | & | & & | \end{bmatrix}\\ &=\begin{bmatrix} c_{1,1} & c_{2,1} & \ldots & c_{d,1}\\ c_{1,2} & c_{2,2} & \ldots & c_{d,2}\\ \vdots & \vdots & & \vdots & & & B & & &\\ c_{1,d} & c_{2,d} & \ldots & c_{d,d}\\ 0 & 0 & \ldots & 0\\ 0 & 0 & \ldots & 0\\ \vdots & \vdots & & \vdots & & & D & & &\\ 0 & 0 & \ldots & 0 \end{bmatrix}\\ &= \begin{bmatrix} C & B\\ O_{n-d,d} & D \end{bmatrix}. \end{aligned} $$ 其中 $B,D$ 是兩個不重要的矩陣. :::success Nice~ ::: ##### Exercise 7(b) 若 $A$ 為一矩陣,而 $\lambda$ 為其一特徵值。 說明 $\ker(A - \lambda I)$ 為一 $A$-不變子空間。 **Ans:** 對於所有的 $\bx \in \ker(A - \lambda I)$,都有 $(A - \lambda I) \bx = \bzero$,則可以有以下推論 $$ \begin{aligned} &(A - \lambda I) \bx = \bzero\\ \implies & A\bx - \lambda \bx = \bzero\\ \implies & AA\bx - A\lambda \bx = A\bzero = \bzero\\ \implies & A(A\bx - \lambda \bx) = \bzero\\ \implies & AA\bx - A \lambda \bx = \bzero\\ \implies & AA\bx - \lambda A \bx = \bzero\\ \implies & (A - \lambda I)A\bx = \bzero\\ \implies & A\bx \in \ker(A-\lambda I). \end{aligned} $$ :::info 最後這題感覺有點繞路,不過證明沒問題。 ::: :::info 目前分數 = 6.5 &times; 檢討 = 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