Matrix Spaces
A Matrix space is a Vector Space where elements are matrices
E.g. Space $M$ - $3 \times 3$ matrices
- any $3 \times 3$ matrix an element of this space $M$ (“vector” in $M$)
- we can multiply by a scalar and add two matrices - which is why we can call it “vector space”
Subspaces
Subspaces of the matrix space should form a space on their own.
- What are subspaces of the matrix space?
- All upper-triangular matrices
- all symmetric matrices
- diagonal matrices (upper-triangular $\cup$ symmetric)
Bases
What about bases for such spaces?
E.g. $M$: $3 \times 3$ matrices:
- $\begin{bmatrix}
1 & 0 & 0 \
0 & 0 & 0 \
0 & 0 & 0
\end{bmatrix}, \begin{bmatrix} 0 & 1 & 0 \ 0 & 0 & 0 \ 0 & 0 & 0
\end{bmatrix}, \begin{bmatrix} 0 & 0 & 1 \ 0 & 0 & 0 \ 0 & 0 & 0
\end{bmatrix}, … , \begin{bmatrix} 0 & 0 & 0 \ 0 & 0 & 0 \ 0 & 0 & 1
\end{bmatrix}$ - $\text{dim}\big( M \big) = 9$
$S$ - subspace of $M$, symmetric $3 \times 3$ matrices
- $\text{dim}\big( S \big) = 6$ - because only 6 elements change in this subspace
$U$ - subspace of $M$ with upper-diagonal matrices
- $\text{dim}\big( U\big) = 6$ as well - same reason (but have zeros for the upper corner)
$S \cup U$ - symmetric and upper-diagonal $\Rightarrow$ diagonal matrices
- $\text{dim}\big( S \cup U\big) = 3$
$S \cap U$
- not a subspace:
- $S$ is 6-dim, $U$ is 3-dim
$S + U$
- any matrix from $S$ plus any matrix from $U$
- this way we can get possible matrix
- so it’s also a subspace
- $\text{dim}\big( S + U\big) = 9$
rule:
- $\text{dim}\big( S \big) + \text{dim}\big( U \big) = \text{dim}\big( S \cap U \big) + \text{dim}\big( S + U \big)$
Inner Product
How do we define the inner product?
- Element-wise: $\langle A, B \rangle = \sum_{ij} a_{ij} b_{ij}$
-
then the norm based on this product is $| A |_F = \langle A, A \rangle$, it’s called the Frobenius Norm.
Sources
- Linear Algebra MIT 18.06 (OCW)
- Kalman, Dan. “A singularly valuable decomposition: the SVD of a matrix.” (1996). [http://www.math.washington.edu/~morrow/498_13/svd.pdf]