An examination of different fitness and novelty based selection methods for the evolution of neural networks

作     者：Inden, Benjamin Jin, Yaochu Haschke, Robert Ritter, Helge Sendhoff, Bernhard 

作者机构：Univ Bielefeld Res Inst Cognit & Robot D-33615 Bielefeld Germany Univ Surrey Dept Comp Guildford GU2 5XH Surrey England Univ Bielefeld Neuroinformat Grp D-33615 Bielefeld Germany Honda Res Inst Europe Offenbach Germany 

出 版 物：《SOFT COMPUTING》 (Soft Comput.)

年 卷 期：2013年第17卷第5期

页      面：753-767页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Honda Research Institute Europe 

主　　题：Neuroevolution Selection Novelty search Evolutionary robotics NEAT 

摘      要：It has been suggested recently that it is a reasonable abstraction of evolutionary processes to use evolutionary algorithms that select individuals based on the novelty of their behavior instead of their fitness. Here we study the performance of fitness- and novelty-based search on several neuroevolution tasks. We also propose several new algorithms that select both for fit and for novel individuals, but without weighting these two criteria directly against each other. We find that behavioral speciation, behavioral near neutral speciation, and behavioral novelty speciation perform best on most tasks. Pure novelty search, as well as a number of hybrid methods without speciation mechanism, do not perform well on most tasks. Using behavioral criteria for speciation often yields better results than using genetic criteria.