Diagonal scalings for the eigenstructure of arbitrary pencils

by   Froilán M. Dopico, et al.

In this paper we show how to construct diagonal scalings for arbitrary matrix pencils λ B-A, in which both A and B are complex matrices (square or nonsquare). The goal of such diagonal scalings is to "balance" in some sense the row and column norms of the pencil. We see that the problem of scaling a matrix pencil is equivalent to the problem of scaling the row and column sums of a particular nonnegative matrix. However, it is known that there exist square and nonsquare nonnegative matrices that can not be scaled arbitrarily. To address this issue, we consider an approximate embedded problem, in which the corresponding nonnegative matrix is square and can always be scaled. The new scaling method is then based on the Sinkhorn-Knopp algorithm for scaling a square nonnegative matrix with total support to be doubly stochastic. In addition, using results of U. G. Rothblum and H. Schneider (1989), we give sufficient conditions for the existence of diagonal scalings of square nonnegative matrices to be not only doubly stochastic but have any prescribed common vector for the row and column sums. We illustrate numerically that the new scaling techniques for pencils improve the sensitivity of the computation of their eigenvalues.



There are no comments yet.


page 1

page 2

page 3

page 4


Sinkhorn limits in finitely many steps

Applied to a nonnegative m× n matrix with a nonzero σ-diagonal, the sequ...

Operator scaling with specified marginals

The completely positive operators, which can be viewed as a generalizati...

Scaling positive random matrices: concentration and asymptotic convergence

It is well known that any positive matrix can be scaled to have prescrib...

Finding Hall blockers by matrix scaling

For a given nonnegative matrix A=(A_ij), the matrix scaling problem asks...

Identification of Matrix Joint Block Diagonalization

Given a set 𝒞={C_i}_i=1^m of square matrices, the matrix blind joint blo...

Sparse Matrix Inversion with Scaled Lasso

We propose a new method of learning a sparse nonnegative-definite target...

Robust and effective eSIF preconditioning for general SPD matrices

We propose an unconditionally robust and highly effective preconditioner...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.