A fast algorithm for matrix balancing

Knight, Philip and Ruiz, Daniel (2013) A fast algorithm for matrix balancing. IMA Journal of Numerical Analysis, 33 (3). pp. 1029-1047. ISSN 0272-4979 (https://doi.org/10.1093/imanum/drs019)

[thumbnail of A fast algorithm for matrix balancing]
Preview
 PDF. Filename: fastbalance.pdf
Preprint
Download (223kB)| Preview

Abstract

As long as a square nonnegative matrix A contains sufficient nonzero elements, then the matrix can be balanced, that is we can find a diagonal scaling of A that is doubly stochastic. A number of algorithms have been proposed to achieve the balancing, the most well known of these being Sinkhorn-Knopp. In this paper we derive new algorithms based on inner-outer iteration schemes. We show that Sinkhorn-Knopp belongs to this family, but other members can converge much more quickly. In particular, we show that while stationary iterative methods offer little or no improvement in many cases, a scheme using a preconditioned conjugate gradient method as the inner iteration can give quadratic convergence at low cost.

CORE (COnnecting REpositories)