Title :
An Efficient Pareto-Coevolution Archive
Author :
Yang, Liping ; Huang, Houkuan ; Yang, Xiaohong
Author_Institution :
Beijing Jiaotong Univ., Beijing
Abstract :
Archive-based methods have been proposed for guaranteeing progress in coevolution. An important question in the practical applications is how to minimize the size of archives while remaining reliability. In the paper we present an efficient Pareto- Coevolution archive based on the problem structure which is identified in coevolution process. In the design, the size of test-archive is minimized while standing for currently global progress. Compared with IPCA on a Numbers game problem, experimental results show that our algorithm is more efficient.
Keywords :
Pareto optimisation; game theory; Pareto-coevolution archive; numbers game problem; problem structure; Application software; Finance; Function approximation; Information technology; Performance evaluation; Process design; Reliability engineering; Testing;
Conference_Titel :
Natural Computation, 2007. ICNC 2007. Third International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2875-5
DOI :
10.1109/ICNC.2007.193
