Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for $\ell_1$-Norm Principal Component Analysis

TL;DR

This paper introduces a proximal alternating minimization method with extrapolation (PAMe) for solving the challenging non-smooth, non-convex L1-PCA problem, establishing its linear convergence and demonstrating its effectiveness through experiments.

Contribution

The paper develops a new algorithm, PAMe, with proven linear convergence for L1-PCA, advancing theoretical understanding and practical solution methods.

Findings

PAMe achieves linear convergence for L1-PCA.

Numerical experiments show PAMe is competitive with existing methods.

Theoretical analysis reveals the Kurdyka-ojasiewicz exponent at critical points is 1/2.

Abstract

A popular robust alternative of the classic principal component analysis (PCA) is the $\ell_1$-norm PCA (L1-PCA), which aims to find a subspace that captures the most variation in a dataset as measured by the $\ell_1$-norm. L1-PCA has shown great promise in alleviating the effect of outliers in data analytic applications. However, it gives rise to a challenging non-smooth non-convex optimization problem, for which existing algorithms are either not scalable or lack strong theoretical guarantees on their convergence behavior. In this paper, we propose a proximal alternating minimization method with extrapolation (PAMe) for solving a two-block reformulation of the L1-PCA problem. We then show that for both the L1-PCA problem and its two-block reformulation, the Kurdyka-\L ojasiewicz exponent at any of the limiting critical points is $1/2$. This allows us to establish the linear convergence of the sequence of iterates generated by PAMe and to determine the criticality of the limit of the sequence with respect to both the L1-PCA problem and its two-block reformulation. To complement our theoretical development, we show via numerical experiments on both synthetic and real-world datasets that PAMe is competitive with a host of existing methods. Our results not only significantly advance the convergence theory of iterative methods for L1-PCA but also demonstrate the potential of our proposed method in applications.