| 1 |
From the previous discussion, we see that matrix multiplication is actually very rarely used in scientific computing for large N. (Yet it's a favorite computer science algorithm!)
|
| 2 |
Full matrix equation solvers are sometimes used but not very common. Of course equation solvers are incredibly important for sparse matrices because if the matrix is large
-
the physics of the problem will make it sparse
-
it will be insoluble unless it is sparse
|
| 3 |
In solving , formally , but this is NOT normally the best numerical method.
|
| 4 |
If is sparse, both and the better LU decomposition are NOT sparse.
|
