Title of Paper:A SUBGRADIENT-BASED CONVEX APPROXIMATIONS METHOD FOR DC PROGRAMMING AND ITS APPLICATIONS

Hits:

Date of Publication:2016-10-01

Journal:JOURNAL OF INDUSTRIAL AND MANAGEMENT OPTIMIZATION

Included Journals:SCIE、Scopus

Volume:12

Issue:4

Page Number:1349-1366

ISSN No.:1547-5816

Key Words:DC programming; convex program; stationary point; set-valued analysis; convergence

Abstract:We consider an optimization problem that minimizes a function of the form f = f(0) + f(1) - f(2) with the constraint g-h <= 0, where f(0) is continuous differentiable, f(1), f(2) are convex and g,h are lower semicontinuous convex. We propose to solve the problem by an inexact subgradient-based convex approximations method. Under mild assumptions, we show that the method is guaranteed to converge to a stationary point. Finally, some preliminary numerical results are given.