On the combination of coevolution and novelty search

This paper develops a new method for coevolution, named Fitness-Diversity Driven Coevolution (FDDC). This approach builds on existing methods by a combination of a (predator-prey) Coevolutionary Genetic Algorithm (CGA) and novelty search. The innovation lies in replacing the absolute novelty measure with a relative one, called Fitness-Diversity. FDDC overcomes problems common in both CGAs (premature convergence and unbalanced coevolution) and in novelty search (construction of an archive). As a proof of principle, Spring Loaded Inverted Pendulums (SLIPs) are coevolved with 2D-terrains the SLIPs must learn to traverse.

[1]  Samir W. Mahfoud.  Crowding and Preselection Revisited , 1992, PPSN.

[2]  Jan Paredis,et al.  Coevolutionary Computation , 1995, Artificial Life.

[3]  Katie Byl,et al.  Enforced symmetry of the stance phase for the Spring-Loaded Inverted Pendulum , 2012, 2012 IEEE International Conference on Robotics and Automation.

[4]  David E. Goldberg,et al.  Genetic Algorithms with Sharing for Multimodalfunction Optimization , 1987, ICGA.

[5]  Jan Paredis,et al.  Towards Balanced Coevolution , 2000, PPSN.

[6]  Donald S. Fussell,et al.  Computer rendering of stochastic models , 1982, Commun. ACM.

[7]  Richard K. Belew,et al.  New Methods for Competitive Coevolution , 1997, Evolutionary Computation.

[8]  L. Darrell Whitley,et al.  The GENITOR Algorithm and Selection Pressure: Why Rank-Based Allocation of Reproductive Trials is Best , 1989, ICGA.

[9]  W. Daniel Hillis,et al.  Co-evolving parasites improve simulated evolution as an optimization procedure , 1990 .

[10]  H. Benjamin Brown,et al.  Evidence for Spring Loaded Inverted Pendulum Running in a Hexapod Robot , 2000, ISER.

[11]  Giovanni Squillero,et al.  Divergence of character and premature convergence: A survey of methodologies for promoting diversity in evolutionary optimization , 2016, Inf. Sci..

[12]  Kenneth O. Stanley,et al.  Abandoning Objectives: Evolution Through the Search for Novelty Alone , 2011, Evolutionary Computation.

[13]  K. Dejong,et al.  An analysis of the behavior of a class of genetic adaptive systems , 1975 .

[14]  J. Pollack,et al.  Challenges in coevolutionary learning: arms-race dynamics, open-endedness, and medicocre stable states , 1998 .