向量化和矩阵化

向量化和矩阵化

vectorization
matricization

看论文遇到向量化和矩阵化的概念
We use vec(·), vech(·) and mat(·) for vectorization, half vectorization, and matricization respectively.

Let $e_i$ be the $i$th canonical basis vector of ${\mathbb{R}}^n$ and $E_{ij}=e_i{e}_j^T$ .

The column-wise block matrix $B_i\in {\mathbb{R}}^{n^2\times n}$ consists of n blocks of size $n\times n$, with an identity matrix only in the ith block and the others are all zeros. Then for any matrix $X\in {\mathbb{R}}^{n\times n}$ and vector $x\in {\mathbb{R}}^{n^2}$ , vectorization and matricization can be expressed by linear operators as $$\mathrm{vec}(X)=\sum _{i=1}^n{B}_{i}X{e}_i$$

$$\mathrm{mat}(x)=\sum _{i=1}^n{B}_i^{T}x{e}_i^T$$

可以看出$B_i$是第$i$块为单位矩阵，其他元素为0的块矩阵。直接看公式不是那么明显，为了便于理解写个程序验证一下结果。
我们对矩阵
$$X=\left[\begin{array}{ccc}1& 2& 3\\ 4& 5& 6\\ 7& 8& 9\end{array}\right]$$
进行向量化；得到$$xx=\left[\begin{array}{c}1\\ 4\\ 7\\ 2\\ 5\\ 8\\ 3\\ 6\\ 9\end{array}\right]$$

---

对向量
$$xx=\left[\begin{array}{c}1\\ 2\\ 3\\ 4\\ 5\\ 6\\ 7\\ 8\\ 9\end{array}\right]$$
进行矩阵化，得到

$$X=\left[\begin{array}{ccc}1& 4& 7\\ 2& 5& 8\\ 3& 6& 9\end{array}\right]$$

验证代码(Matlab)

B1=[    1 0 0;
    0 1 0;
    0 0 1;
    0  0 0;
    0 0 0;
    0 0 0;
    0 0 0;
    0 0 0;
    0 0 0];
B2=[0  0 0;
    0 0 0;
    0 0 0;
    1 0 0;
    0 1 0;
    0 0 1;
    0 0 0;
    0 0 0;
    0 0 0];
B3=[0  0 0;
    0 0 0;
    0 0 0;
    0 0 0;
    0 0 0;
    0 0 0;
    1 0 0;
    0 1 0;
    0 0 1];
e1 = [1 0 0]';
e2 = [0 1 0]';
e3 = [0 0 1]';
xx = [1 2 3 4 5 6 7 8 9]';  %%%向量x
X = [1 2 3;4 5 6; 7 8 9];   %%%矩阵X
A = B1*X*e1+B2*X*e2+B3*X*e3;  %%把矩阵向量化
AA = B1'*xx*e1'+B2'*xx*e2'+B3'*xx*e3';  %%把向量单位化

向量化的matlab实现

 A = [1,2;3,4];
 B=A(:);
 c=reshape(A,[],1);

$B=\left[\begin{array}{c}1\\ 3\\ 2\\ 4\end{array}\right]$

$c=\left[\begin{array}{c}1\\ 3\\ 2\\ 4\end{array}\right]$