Nonsmooth exact penalty methods for equality-constrained optimization: complexity and implementation

TL;DR

This paper develops an efficient proximal algorithm for nonsmooth exact penalty methods in equality-constrained optimization, providing convergence analysis, complexity bounds, and empirical comparisons with existing solvers.

Contribution

It introduces a proximal-type implementation for nonsmooth exact penalty methods, analyzes their convergence and complexity, and demonstrates practical efficiency over traditional approaches.

Findings

Proposed a proximal algorithm for nonsmooth exact penalty methods.

Established a worst-case complexity bound of O(ε^{-2}) under certain conditions.

Numerical experiments show the method's robustness and competitiveness.

Abstract

Penalty methods are a well known class of algorithms for constrained optimization. They transform a constrained problem into a sequence of unconstrained \emph{penalized} problems in the hope that approximate solutions of the latter converge to a solution of the former. If Lagrange multipliers exist, exact penalty methods ensure that the penalty parameter only need increase a finite number of times, but are typically scorned in smooth optimization for the penalized problems are not smooth. This led researchers to consider the implementation of exact penalty methods inconvenient. Recent advances in proximal methods have led to increasingly efficient solvers for nonsmooth optimization. We study a general exact penalty algorithm and use it to show that the exact $\ell_2$-penalty method for equality-constrained optimization can, in fact, be implemented efficiently by solving the penalized problem using a proximal-type algorithm. We study the convergence of our algorithm and establish a worst-case complexity bound of $\mathcal{O}(\epsilon^{-2})$ to bring a stationarity measure below $\epsilon > 0$ under the Mangarasian-Fromowitz constraint qualification and Lipschitz continuity of the objective gradient and constraint Jacobian. While the Lipschitz continuity of the objective gradient is not required for convergence in view of recent works, it is used in our analysis to derive the complexity bound. In a degenerate scenario where the penalty parameter grows unbounded, the complexity becomes $\mathcal{O}(\epsilon^{-8})$, which is worse than another bound found in the literature. Finally, we report numerical experience on small-scale problems from a standard collection and compare our solver with an augmented-Lagrangian and an SQP method. Our preliminary implementation is superior to the augmented Lagrangian in terms of robustness and efficiency, and is competitive with the SQP method.