Adaptation of Length in a Nonstationary Environment

In this paper, we examine the behavior of a variable length GA in a nonstationary problem environment. Results indicate that a variable length GA is better able to adapt to changes than a fixed length GA. Closer examination of the evolutionary dynamics reveals that a variable length GA can in fact take advantage of its variable length representation to exploit good quality building blocks after a change in the problem environment.

• Type: Conference
• Year: 2003
• Where: GECCO
• Authors: Han Yu, Annie S. Wu, Kuo-Chi Lin, Guy A. Schiavone
• Keywords: GECCO 2003, Problem Environment, Variable Length Ga, Variable Length Representation