User Tools

Site Tools


Plugin installed incorrectly. Rename plugin directory '_include' to 'include'.
Plugin installed incorrectly. Rename plugin directory '__include' to 'include'.
lecture_8_slides

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
lecture_8_slides [2016/02/10 15:37] rupertlecture_8_slides [2017/02/16 08:56] (current) – [Commuting matrices III] rupert
Line 1: Line 1:
-==== Matrix multiplication ====+~~REVEAL~~ 
 + 
 +==== Row-column & matrix multiplication ==== 
 + 
 +  * The **row-column product** of $a$ and $b$ is defined by \[\!\!\!\!\!\!\!\!\!\!ab=[\begin{smallmatrix}a_1&a_2&\dots&a_n\end{smallmatrix}]\left[\begin{smallmatrix}b_1\\b_2\\\vdots\\b_n\end{smallmatrix}\right]=a_1b_1+a_2b_2+\dots+a_nb_n.\] 
 + 
 +  * $AB=$ matrix of all "row-of-$A$ times col-of-$B$" products 
 +  * \[\!\!\!\!\!\!\!\!\!\!\!\!\!\!\! \def\r{\left[\begin{smallmatrix}1&0&5\end{smallmatrix}\right]}\def\rr{\left[\begin{smallmatrix}2&-1&3\end{smallmatrix}\right]}\left[\begin{smallmatrix}1&0&5\\2&-1&3\end{smallmatrix}\right]\left[\begin{smallmatrix} 1&2\\3&4\\5&6\end{smallmatrix}\right]\def\s{\left[\begin{smallmatrix}1\\3\\5\end{smallmatrix}\right]}\def\ss{\left[\begin{smallmatrix}2\\4\\6\end{smallmatrix}\right]}=\left[\begin{smallmatrix}{\r\s}&{\r\ss}\\{\rr\s}&{\rr\ss}\end{smallmatrix}\right]=\left[\begin{smallmatrix}26&32\\14&18\end{smallmatrix}\right].\] 
 + 
 +==== Matrix multiplication: the definition ====
  
   * Let $A,B$ be matrices, with sizes   * Let $A,B$ be matrices, with sizes
Line 23: Line 32:
  
 ==== Example 2 ==== ==== Example 2 ====
-If $A=\mat{1&2\\3&4\\5&6}$, $B=\mat{2&1&1\\1&2&0\\1&0&2\\1&0&2\\2&2&1}$ and $C=\mat{1&3&0&7\\0&4&6&8}$, +If $A=\mat{1&2\\3&4\\5&6}$, $B=\mat{2&1&1\\1&2&0\\1&0&2\\2&2&1}$ and $C=\mat{1&3&0&7\\0&4&6&8}$, 
   * $A$ is $3\times 2$, $B$ is $4\times 3$ and $C$ is $2\times 4$, so    * $A$ is $3\times 2$, $B$ is $4\times 3$ and $C$ is $2\times 4$, so 
   * $AB$, $CA$ and $BC$ don't exist (undefined);   * $AB$, $CA$ and $BC$ don't exist (undefined);
Line 60: Line 69:
 {{page>commute}} {{page>commute}}
  
-  * Because it'not true in general that $AB=BA$, we say that **matrix multiplication is not commutative**.+  * Because it's true that $AB=BA$ for every choice of matrices $A$ and $B$, we say that **matrix multiplication is not commutative**.
  
 ==== Commuting matrices II ==== ==== Commuting matrices II ====
  
-  * What can we say about commuting matrices?  +  * What can we say about a pair of commuting matrices?
   * Suppose $AB=BA$ and think about sizes.   * Suppose $AB=BA$ and think about sizes.
     * $A$: $n\times m$     * $A$: $n\times m$
Line 75: Line 84:
  
   * If $A$ and $B$ commute, they must be square matrices of the same size.   * If $A$ and $B$ commute, they must be square matrices of the same size.
-  * **Some** square matrices $A$ and $B$ of the same size commute...+  * **Some** pairs of square matrices $A$ and $B$ of the same size do commute...
   * ....but not all!    * ....but not all! 
   * See examples above.   * See examples above.
Line 95: Line 104:
   - $I_nB=B=BI_n$ for any $n\times n$ matrix $B$.    - $I_nB=B=BI_n$ for any $n\times n$ matrix $B$. 
     * In particular, $I_n$ commutes with every other square $n\times n$ matrix $B$.     * In particular, $I_n$ commutes with every other square $n\times n$ matrix $B$.
 +
 +==== Proof that $I_nA=A$ for $A$: $n\times m$ ====
 +
 +  * $I_nA$ is $n\times m$ (from definition of matrix multiplication)
 +  * So $I_nA$ has same size as $A$
 +  * $\text{row}_i(I_n)=[0~0~\dots~0~1~0~\dots~0]$, with $1$ in $i$th place
 +  * $\text{col}_j(A)=\mat{a_{1j}\\a_{2j}\\\vdots\\a_{nj}}$
 +  * So $(i,j)$ entry of $I_nA$ is \[\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\text{row}_i(I_n)\cdot \text{col}_j(A)= 0a_{1j}+0a_{2j}+\dots+1a_{ij}+\dots+0a_{nj} =a_{ij}\]
 +  * same as $(i,j)$ entry of $A$.
 +  * So $I_nA=A$
 +
 +==== More proofs====
 +
 +  * Proof that $A=AI_m$ for $A$: $n\times m$ is very similar (exercise)
 +  * Now if $B$ is $n\times n$, take $n=m$ and $A=B$ above:
 +    * $I_nB=B$ and $BI_n=B$
 +    * So $I_nB=B=BI_n$
 +    * So $I_n$ commutes with $B$, for any $n\times n$ matrix $B$.
 +
 +
 +
lecture_8_slides.1455118658.txt.gz · Last modified: by rupert

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki