RE: Re: A matrix decomposition
CleveAshcraft <[email protected]>
| Newsgroups | gmane.comp.mathematics.csc |
|---|---|
| Organization | LSTC |
| Message-ID | <[email protected]> |
Alex> Consider the bipartite graph of A, then your first problem Alex> is to partition the edges of A into the fewest number of Alex> maximum matchings in the graph. Not quite, I think. Disjoint matching perhaps. Here is my present algorithm. k = 0 ; while A != 0 k = k + 1 find maximum matching of A and form A_k A := A - A_k end It should be easy to find a counterexample where this does not find the minimum number of A_k terms.