A Fast Algorithm That Makes Matrices Optimally Sparse.
Abstract
Under a non-degeneracy assumption on the non-zero entries of a given sparse matrix, a polynomially-bounded algorithm is presented that performs row operations on the given matrix which reduce it to a sparsest possible matrix with the same row space. For each row of the matrix, the algorithm performs a maximum cardinality matching on the bipartite graph associated with a submatrix which is induced by that row. The dual of the optimal matching then species the row operations that will be performed on that row. Also described is a variant algorithm that processes the matrix in place, thus conserving storage and time. The modifications needed to apply the algorithm to matrices that do not necessarily satisfy the non-degeneracy assumption are also described. A particularly promising application of this algorithm is in the reduction of linear constraint matrices. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1982
- Accession Number
- ADA120902
Entities
People
- Alan J. Hoffman
- S. Thomas Mccormick
Organizations
- Stanford University