“矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授。
PDF格式学习笔记下载(Academia.edu)
第7章课程讲义下载(PDF)

Summary

  • For a nonsingular matrix $[A]$ on which one can always write it as $$[A] = [L][U]$$ where $[L]$ is a lower triangular matrix, $[U]$ is a upper triangular matrix.
  • Note that not all matrices have LU decomposition, such as $\begin{bmatrix}0& 2\\ 2& 0\end{bmatrix}$. $$\begin{bmatrix}0& 2\\ 2& 0\end{bmatrix}=\begin{bmatrix}1& 0\\ a& 1\end{bmatrix} \begin{bmatrix}b& c\\ 0& d\end{bmatrix} \Rightarrow \begin{cases} b=0\\ ab=2\end{cases}$$ which is contradiction.
  • If one is solving a set of equations $$[A][X]=[B]$$ then $$LUX=B$$ $$\Rightarrow L^{-1}LUX=L^{-1}B$$ $$\Rightarrow UX=L^{-1}B=Y$$ then we have $$\begin{cases}LY=B\\ UX=Y\end{cases}$$ So we can solve the first equation for $[Y]$by using forward substitution and then use the second equation to calculate the solution vector $[X]$ by back substitution.
  • For instance, solve the following set of equations: $$\begin{bmatrix}1& 2& 3\\ 2& 1& -4\\ 1& 5& 2\end{bmatrix}\cdot \begin{bmatrix} x\\ y\\ z \end{bmatrix} = \begin{bmatrix} 14\\ -8\\ 17\end{bmatrix}$$ Applying LU decomposition on the coefficient matrix,
    • Firstly write down an identity matrix (the same size as the coefficient matrix) on the left and the coefficient matrix on the right. $$L\leftarrow\begin{bmatrix}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1 \end{bmatrix} \begin{bmatrix}1& 2& 3\\ 2& 1& -4\\ 1& 5& 2\end{bmatrix}\rightarrow U$$
    • Then applying elementary row operation on the right while simultaneously updating successive columns of the matrix on the left. For example, if we are doing $R_1 + m R_2$ on the right then we will do $C_2-mC_1$ on the left. That is, we will keep the equivalent of the product. $$\begin{bmatrix}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1 \end{bmatrix} \begin{bmatrix}1& 2& 3\\ 2& 1& -4\\ 1& 5& 2\end{bmatrix}$$ $$\Rightarrow\begin{cases}R_2-2R_1 \\ C_1+2C_2\end{cases} \begin{bmatrix}1& 0& 0\\ 2& 1& 0\\ 0& 0& 1 \end{bmatrix} \begin{bmatrix}1& 2& 3\\ 0& -3& -10\\ 1& 5& 2\end{bmatrix}$$ $$\Rightarrow\begin{cases}R_3-R_1 \\ C_1+C_3\end{cases} \begin{bmatrix}1& 0& 0\\ 2& 1& 0\\ 1& 0& 1 \end{bmatrix} \begin{bmatrix}1& 2& 3\\ 0& -3& -10\\ 0& 3& -1\end{bmatrix}$$ $$\Rightarrow\begin{cases}R_3+R_2 \\ C_2-C_3\end{cases} \begin{bmatrix}1& 0& 0\\ 2& 1& 0\\ 1& -1& 1 \end{bmatrix} \begin{bmatrix}1& 2& 3\\ 0& -3& -10\\ 0& 0& -11\end{bmatrix}$$ Thus far, the right matrix is an upper triangular matrix (i.e. $U$) and the left one is a lower triangular matrix (i.e. $L$).
    • Solving $[L][Y]=[B]$, that is $$\begin{bmatrix}1& 0& 0\\ 2& 1& 0\\ 1& -1& 1 \end{bmatrix}\cdot Y=\begin{bmatrix} 14\\ -8\\ 17\end{bmatrix}\Rightarrow Y=\begin{bmatrix}14\\ -36\\ -33\end{bmatrix}$$
    • Solving $[U][X]=[Y]$, that is $$\begin{bmatrix}1& 2& 3\\ 0& -3& -10\\ 0& 0& -11\end{bmatrix}\cdot \begin{bmatrix} x\\ y\\ z \end{bmatrix} = \begin{bmatrix}14\\ -36\\ -33\end{bmatrix}$$ $$ \Rightarrow\begin{cases}x=1\\ y=2 \\ z=3\end{cases}$$

Selected Problems

1. Find the $[L]$ and $[U]$ matrices of the following matrix $$\begin{bmatrix}25& 5& 4\\ 75& 7& 16\\ 12.5& 12& 22 \end{bmatrix}$$

Solution:
$$\begin{bmatrix}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1 \end{bmatrix}\begin{bmatrix}25& 5& 4\\ 75& 7& 16\\ 12.5& 12& 22 \end{bmatrix}$$ $$\Rightarrow \begin{cases}R_2-3R_1\\ R_3-{1\over2}R_1\\ C_1+3C_2\\ C_1+{1\over2}C_3\end{cases} \begin{bmatrix}1& 0& 0\\ 3& 1& 0\\ {1\over2}& 0& 1 \end{bmatrix} \begin{bmatrix}25& 5& 4\\ 0& -8& 4\\ 0& 9.5& 20 \end{bmatrix}$$ $$\Rightarrow \begin{cases}R_3+{19\over16}R_2\\C_2-{19\over16}C_3\end{cases} \begin{bmatrix}1& 0& 0\\ 3& 1& 0\\ {1\over2}& -{19\over16}& 1 \end{bmatrix} \begin{bmatrix}25& 5& 4\\ 0& -8& 4\\ 0& 0& {99\over4} \end{bmatrix}$$ That is, $$L= \begin{bmatrix}1& 0& 0\\ 3& 1& 0\\ {1\over2}& -{19\over16}& 1 \end{bmatrix},\ U = \begin{bmatrix}25& 5& 4\\ 0& -8& 4\\ 0& 0& {99\over4} \end{bmatrix}.$$

2. Using LU decomposition to solve: $$\begin{cases} 4x_1 + x_2 - x_3 = -2\\ 5x_1+x_2+2x_3=4\\ 6x_1+x_2+x_3=6 \end{cases}$$

Solution:
$$\begin{bmatrix}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1 \end{bmatrix} \begin{bmatrix}4& 1& -1\\ 5& 1& 2\\ 6& 1& 1\end{bmatrix}$$ $$\Rightarrow \begin{cases}R_2-{5\over4}R_1\\ R_3-{3\over2}R_1\\ C_1+{5\over4}C_2\\ C_1+{3\over2}C_3\end{cases} \begin{bmatrix}1& 0& 0\\ {5\over4}& 1& 0\\ {3\over2}& 0& 1 \end{bmatrix} \begin{bmatrix}4& 1& -1\\ 0& -{1\over4}& {13\over4}\\ 0& -{1\over2}& {5\over2}\end{bmatrix}$$ $$\Rightarrow \begin{cases}R_3-2R_2\\ C_2+2C_3\end{cases} \begin{bmatrix}1& 0& 0\\ {5\over4}& 1& 0\\ {3\over2}& 2& 1 \end{bmatrix} \begin{bmatrix}4& 1& -1\\ 0& -{1\over4}& {13\over4}\\ 0&0& -4\end{bmatrix}$$ That is, $$L = \begin{bmatrix}1& 0& 0\\ {5\over4}& 1& 0\\ {3\over2}& 2& 1 \end{bmatrix},\ U= \begin{bmatrix}4& 1& -1\\ 0& -{1\over4}& {13\over4}\\ 0&0& -4\end{bmatrix}.$$ Then we solve $[L][Y]=[B]$, $$\begin{bmatrix}1& 0& 0\\ {5\over4}& 1& 0\\ {3\over2}& 2& 1 \end{bmatrix}\cdot Y=\begin{bmatrix}-2\\4\\6\end{bmatrix} \Rightarrow Y=\begin{bmatrix}-2\\{13\over2}\\ -4\end{bmatrix}$$ Finally, we solve $[U][X]=[Y]$, $$\begin{bmatrix}4& 1& -1\\ 0& -{1\over4}& {13\over4}\\ 0&0& -4\end{bmatrix}\cdot X= \begin{bmatrix}-2\\{13\over2}\\ -4\end{bmatrix}\Rightarrow X=\begin{bmatrix}3\\-13\\1\end{bmatrix}$$ Thus the solution is $$\begin{cases}x_1 = 3\\ x_2 = -13\\ x_3 = 1\end{cases}$$

3. Find the inverse of $$[A]=\begin{bmatrix}3& 4& 1\\ 2& -7& -1\\ 8& 1& 5\end{bmatrix}$$

Solution:

To find the inverse of a matrix, actually it is to solve a set of equations: $$\begin{cases}AX_1=[1, 0, 0]^{T}\\ AX_2 = [0, 1, 0]^{T}\\ AX_3 = [0, 0, 1]^{T} \end{cases}$$ Firstly, we will find the $[L]$ and $[U]$. $$\begin{bmatrix}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1 \end{bmatrix} \begin{bmatrix}3& 4& 1\\ 2& -7& -1\\ 8& 1& 5\end{bmatrix}$$ $$\Rightarrow \begin{cases}R_2-{2\over3}R_1\\ R_3-{8\over3}R_1\\ C_1+{2\over3}C_2\\ C_1+{8\over3}C_3\end{cases} \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 0& 1 \end{bmatrix} \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0& -{29\over3}& {7\over3}\end{bmatrix}$$ $$\Rightarrow \begin{cases}R_3-R_2\\ C_2+C_3\end{cases} \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 1& 1 \end{bmatrix} \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0&0& 4\end{bmatrix}$$ That is, $$L = \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 1& 1 \end{bmatrix},\ U= \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0&0& 4\end{bmatrix}.$$ Then we solve $[L][Y]=[I]$, note that there are three columns of $[Y]$: $$LY_1 = \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 1& 1 \end{bmatrix} \cdot Y_1 = \begin{bmatrix}1\\0\\0\end{bmatrix} \Rightarrow Y_1=\left[1, -{2\over3}, -2\right]^{T}$$ $$LY_2 = \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 1& 1 \end{bmatrix} \cdot Y_2 = \begin{bmatrix}0\\1\\0\end{bmatrix} \Rightarrow Y_2=\left[0, 1, -1\right]^{T}$$ $$LY_3 = \begin{bmatrix}1& 0& 0\\ {2\over3}& 1& 0\\ {8\over3}& 1& 1 \end{bmatrix} \cdot Y_3 = \begin{bmatrix}0\\0\\1\end{bmatrix} \Rightarrow Y_3=\left[0, 0, 1\right]^{T}$$ Finally we can solve $[X]$ by $[U][X]=[Y]$: $$UX_1=Y_1\Rightarrow \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0&0& 4\end{bmatrix} \cdot X_1 = \begin{bmatrix}1\\ -{2\over3}\\ -2\end{bmatrix}\Rightarrow X_1= \left[{17\over58}, {9\over58}, -{1\over2}\right]^{T}$$ $$UX_2=Y_2\Rightarrow \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0&0& 4\end{bmatrix} \cdot X_2 = \begin{bmatrix}0\\ 1\\ -1\end{bmatrix}\Rightarrow X_2= \left[{19\over116}, -{7\over116}, -{1\over4}\right]^{T}$$ $$UX_3=Y_3\Rightarrow \begin{bmatrix}3& 4& 1\\ 0& -{29\over3}& -{5\over3}\\ 0&0& 4\end{bmatrix} \cdot X_3 = \begin{bmatrix}0\\ 0\\ 1\end{bmatrix}\Rightarrow X_3= \left[-{3\over116}, -{5\over116}, {1\over4}\right]^{T}$$ Thus the inverse of the original matrix is $$[A]^{-1} = \begin{bmatrix}{17\over58} & {19\over116} & -{3\over116}\\ {9\over58} & -{7\over116} & -{5\over116}\\ -{1\over2} & -{1\over4} & {1\over4}\end{bmatrix}$$

A.Kaw矩阵代数初步学习笔记 7. LU Decomposition的更多相关文章

  1. A.Kaw矩阵代数初步学习笔记 10. Eigenvalues and Eigenvectors

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  2. A.Kaw矩阵代数初步学习笔记 9. Adequacy of Solutions

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  3. A.Kaw矩阵代数初步学习笔记 8. Gauss-Seidel Method

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  4. A.Kaw矩阵代数初步学习笔记 6. Gaussian Elimination

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  5. A.Kaw矩阵代数初步学习笔记 5. System of Equations

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  6. A.Kaw矩阵代数初步学习笔记 4. Unary Matrix Operations

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  7. A.Kaw矩阵代数初步学习笔记 3. Binary Matrix Operations

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  8. A.Kaw矩阵代数初步学习笔记 2. Vectors

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

  9. A.Kaw矩阵代数初步学习笔记 1. Introduction

    “矩阵代数初步”(Introduction to MATRIX ALGEBRA)课程由Prof. A.K.Kaw(University of South Florida)设计并讲授. PDF格式学习笔 ...

随机推荐

  1. 【转】如何拿到半数面试公司Offer——我的Python求职之路

    原文地址 从八月底开始找工作,短短的一星期多一些,面试了9家公司,拿到5份Offer,可能是因为我所面试的公司都是些创业性的公司吧,不过还是感触良多,因为学习Python的时间还很短,没想到还算比较容 ...

  2. js的单引号,双引号,转移符

    这里我们看到想在style后边在插入一个样式的变量,data.cssSytle.a是做边和邮编都是"",并且没有转移符

  3. Graphql介绍(Introduction to GraphQL)

    Introduction to GraphQL  GraphQL介绍 Learn about GraphQL, how it works, and how to use it in this seri ...

  4. yii2干货

    Sites 网站 yiifeed:Yii 最新动态都在这里 yiigist:Yii 专用的 Packages my-yii:Yii 学习资料和新闻 Docs 文档 Yii Framework 2.0 ...

  5. android 使用多个接口

    今天,好久没有这么用过都忘记可以这样用了.来记录下: 一个类想要使用多个接口可以implements 接口1 , 接口2,...

  6. SharePoint Backup

    这里主要介绍使用admin center直接backup: 1.浏览器进入管理中心,选择备份: 2.按需要选择需要备份的内容 3.选择备份位置,然后等待服务器备份完成(windows explore中 ...

  7. 2016年GitHub 排名前 100 的安卓、iOS项目简介(收藏)

    排名完全是根据 GitHub 搜索 Java 语言选择 (Best Match) 得到的结果, 然后过滤了跟 Android 不相关的项目, 所以排名并不具备任何官方效力, 仅供参考学习, 方便初学者 ...

  8. 函数也是对象,本片介绍函数的属性、方法、Function()狗仔函数。

    1.arguments.length表示实参的个数. 2.arguments.callee.length表示形参个数. function test(a,b,c,d,e,f){ alert(argume ...

  9. ScrollView内部元素如何做到fill_parent 或者 match_parent?

    转  : http://www.jcodecraeer.com/a/anzhuokaifa/androidkaifa/2014/0704/1629.html ScrollView滚动视图是指当拥有很多 ...

  10. Android数据格式解析对象JSON用法(转)

    地址:http://www.cnblogs.com/devinzhang/archive/2012/01/09/2317315.html 里面的重点: JSON解析案例     (1)解析Object ...