A new asynchronous parallel evolutionary algorithm for function optimization

Abstract

This paper introduces a new asynchronous parallel evolutionary algorithm (APEA) based on the island model for solving function optimization problems. Our fully distributed APEA overlaps the communication and computation efficiently and is inherently fault-tolerant in a large-scale distributed computing environment. For the scalable BUMP problem, our APEA algorithm achieves the best solution for the 50-dimension problem, and is the first algorithm of which we are aware that can solve the 1,000,000- dimension problem. For other benchmark problems, our APEA finds the best solution to G7 in fewer time steps than [16,17], and finds a better solution to G10 than [17].