Inexact accelerated proximal gradient method with line search and reduced complexity for affine-constrained and bilinear saddle-point structured convex problems

TL;DR

This paper introduces an inexact accelerated proximal gradient method with line search that reduces computational complexity for affine-constrained and bilinear saddle-point structured convex problems by efficiently stopping inner loops based on stationarity measures.

Contribution

The paper proposes a novel inexact APG method with adaptive stopping criteria, improving efficiency over existing gradient sliding techniques for structured convex optimization problems.

Findings

Significantly higher efficiency demonstrated in numerical experiments.

Reduces total complexity by adaptively stopping inner loops.

Applicable to affine-constrained and bilinear saddle-point problems.

Abstract

The goal of this paper is to reduce the total complexity of gradient-based methods for two classes of problems: affine-constrained composite convex optimization and bilinear saddle-point structured non-smooth convex optimization. Our technique is based on a double-loop inexact accelerated proximal gradient (APG) method for minimizing the summation of a non-smooth but proximable convex function and two smooth convex functions with different smoothness constants and computational costs. Compared to the standard APG method, the inexact APG method can reduce the total computation cost if one smooth component has higher computational cost but a smaller smoothness constant than the other. With this property, the inexact APG method can be applied to approximately solve the subproblems of a proximal augmented Lagrangian method for affine-constrained composite convex optimization and the smooth approximation for bilinear saddle-point structured non-smooth convex optimization, where the smooth function with a smaller smoothness constant has significantly higher computational cost. Thus it can reduce total complexity for finding an approximately optimal/stationary solution. This technique is similar to the gradient sliding technique in the literature. The difference is that our inexact APG method can efficiently stop the inner loop by using a computable condition based on a measure of stationarity violation, while the gradient sliding methods need to pre-specify the number of iterations for the inner loop. Numerical experiments demonstrate significantly higher efficiency of our methods over an optimal primal-dual first-order method and the gradient sliding methods.