A min-max reformulation and proximal algorithms for a class of structured nonsmooth fractional optimization problems

TL;DR

This paper introduces a novel min-max reformulation and proximal algorithms for structured nonsmooth fractional optimization problems, enabling effective solutions for applications like sparse signal recovery.

Contribution

It proposes a new reformulation and an alternating maximization proximal descent algorithm with convergence guarantees for a class of nonsmooth fractional problems.

Findings

Algorithm converges to a critical point within O(ε^{-2}) iterations.

Numerical experiments demonstrate the method's efficiency.

Applicable to scale-invariant sparse signal recovery.

Abstract

In this paper, we consider a class of structured nonsmooth fractional minimization, where the first part of the objective is the ratio of a nonnegative nonsmooth nonconvex function to a nonnegative nonsmooth convex function, while the second part is the difference of a smooth nonconvex function and a nonsmooth convex function. This model problem has many important applications, for example, the scale-invariant sparse signal recovery in signal processing. However, the existing methods for fractional programs are not suitable for solving this problem due to its special structure. We first present a novel nonfractional min-max reformulation for the original fractional program and show the connections between their global (local) optimal solutions and stationary points. Based on the reformulation, we propose an alternating maximization proximal descent algorithm and show its subsequential convergence towards a critical point of the original fractional program under a mild assumption. Moreover, we prove that the proposed algorithm can find an $\epsilon$-critical point of the considered problem within $\mathcal{O}(\epsilon^{-2})$ iterations. By further assuming the Kurdyka-{\L}ojasiewicz (KL) property of an auxiliary function, we also establish the convergence of the entire solution sequence generated by the proposed algorithm. Finally, some numerical experiments on the $L_1/L_2$ least squares problem and scale-invariant sparse signal recovery are conducted to demonstrate the efficiency of the proposed method.