Coderivative-Based Semi-Newton Method in Nonsmooth Difference Programming

TL;DR

This paper introduces a novel semi-Newton method based on coderivatives for solving nonsmooth difference-of-functions optimization problems, with proven convergence and practical applications in biochemical and quadratic programming problems.

Contribution

It develops a new Newton-type algorithm utilizing coderivatives for nonsmooth difference programming, with convergence analysis and real-world applications.

Findings

Algorithm demonstrates superior convergence in practical problems.

Effective in biochemical and constrained quadratic programming.

Outperforms some existing techniques in numerical experiments.

Abstract

This paper addresses the study of a new class of nonsmooth optimization problems, where the objective is represented as a difference of two generally nonconvex functions. We propose and develop a novel Newton-type algorithm to solving such problems, which is based on the coderivative generated second-order subdifferential (generalized Hessian) and employs advanced tools of variational analysis. Well-posedness properties of the proposed algorithm are derived under fairly general requirements, while constructive convergence rates are established by using additional assumptions including the Kurdyka--{\L}ojasiewicz condition. We provide applications of the main algorithm to solving a general class of nonsmooth nonconvex problems of structured optimization that encompasses, in particular, optimization problems with explicit constraints. Finally, applications and numerical experiments are given for solving practical problems that arise in biochemical models, constrained quadratic programming, etc., where advantages of our algorithms are demonstrated in comparison with some known techniques and results.