Convergence Analysis for Minimizing Differences of Convex Functions

Nguyen Thai An, Nguyen Mau Nam, Nguyen Dong Yen

Published 2015-04-30Version 1

Several optimization schemes have been known for convex optimization problems over the past several decades. However, numerical algorithms for solving nonconvex optimization problems are still underdeveloped. The first progress to go beyond convexity was made by considering the class of functions are representable as differences of convex functions. One of the most successful algorithms for minimizing this class of functions called the DCA was introduced by Pham Dinh and Le Thi. This algorithm now plays a crucial role in many real-life applications. In this paper, we develop a simple approach to analyze the convergence of the DCA.