Fast Computation of the Rank Profile Matrix and the Generalized Bruhat Decomposition

TL;DR

This paper introduces a new invariant called the rank profile matrix, which captures comprehensive rank profile information of matrices, and presents efficient algorithms for its computation, improving matrix decomposition methods and their practical applications.

Contribution

The paper defines the rank profile matrix over rings, develops algorithms to compute it via PLUQ decompositions, and connects it to Bruhat and related decompositions, enhancing matrix factorization techniques.

Findings

A unique rank profile matrix exists over rings with McCoy's rank.

New algorithms compute the invariant efficiently, especially for low-rank matrices.

Implementation improvements significantly enhance practical matrix decomposition performance.

Abstract

The row (resp. column) rank profile of a matrix describes the stair-case shape of its row (resp. column) echelon form. We here propose a new matrix invariant, the rank profile matrix, summarizing all information on the row and column rank profiles of all the leading sub-matrices. We show that this normal form exists and is unique over any ring, provided that the notion of McCoy's rank is used, in the presence of zero divisors. We then explore the conditions for a Gaussian elimination algorithm to compute all or part of this invariant, through the corresponding PLUQ decomposition. This enlarges the set of known Elimination variants that compute row or column rank profiles. As a consequence a new Crout base case variant significantly improves the practical efficiency of previously known implementations over a finite field. With matrices of very small rank, we also generalize the techniques of Storjohann and Yang to the computation of the rank profile matrix, achieving an $(r^\omega+mn)^{1+o(1)}$ time complexity for an $m \times n$ matrix of rank $r$, where $\omega$ is the exponent of matrix multiplication. Finally, by give connections to the Bruhat decomposition, and several of its variants and generalizations. Thus, our algorithmic improvements for the PLUQ factorization, and their implementations, directly apply to these decompositions. In particular, we show how a PLUQ decomposition revealing the rank profile matrix also reveals both a row and a column echelon form of the input matrix or of any of its leading sub-matrices, by a simple post-processing made of row and column permutations.