Better and Simpler Error Analysis of the Sinkhorn-Knopp Algorithm for Matrix Scaling

TL;DR

This paper provides a simpler, improved convergence analysis of the Sinkhorn-Knopp matrix scaling algorithm using KL-divergence and introduces a new inequality that enhances the understanding of its convergence rate.

Contribution

The authors present an elementary, self-contained convergence analysis for Sinkhorn-Knopp, introducing a new inequality that strengthens convergence guarantees, especially for the -error.

Findings

Improved bounds on the number of iterations for convergence.

Introduction of a new inequality relating KL-divergence to  and  distances.

Enhanced understanding of the -error convergence rate.

Abstract

Given a non-negative $n \times m$ real matrix $A$, the {\em matrix scaling} problem is to determine if it is possible to scale the rows and columns so that each row and each column sums to a specified target value for it. This problem arises in many algorithmic applications, perhaps most notably as a preconditioning step in solving a linear system of equations. One of the most natural and by now classical approach to matrix scaling is the Sinkhorn-Knopp algorithm (also known as the RAS method) where one alternately scales either all rows or all columns to meet the target values. In addition to being extremely simple and natural, another appeal of this procedure is that it easily lends itself to parallelization. A central question is to understand the rate of convergence of the Sinkhorn-Knopp algorithm. In this paper, we present an elementary convergence analysis for the Sinkhorn-Knopp algorithm that improves upon the previous best bound. In a nutshell, our approach is to show a simple bound on the number of iterations needed so that the KL-divergence between the current row-sums and the target row-sums drops below a specified threshold $\delta$, and then connect the KL-divergence with $\ell_1$ and $\ell_2$ distances. For $\ell_1$, we can use Pinsker's inequality. For $\ell_2$, we develop a strengthening of Pinsker's inequality, called (KL vs $\ell_1/\ell_2$) in the paper, which lower bounds the KL-divergence by a combination of $\ell_1$ and $\ell_2$ distance. This inequality may be of independent interest. The idea of studying Sinkhorn-Knopp convergence via KL-divergence is not new and has indeed been previously explored. Our contribution is an elementary, self-contained presentation of this approach and an interesting new inequality that yields a significantly stronger convergence guarantee for the extensively studied $\ell_2$-error.