Neutral networks in an evolutionary robotics search space

Recent work has argued for the importance of non-adaptive neutral evolution in optimisation over difficult search landscapes. We show that the search process underlying a difficult evolutionary robotics problem does indeed show phases of neutral evolution. The noise in evaluated fitness of a single genotype is shown to be able to account for the variance in fitness across a long period of the evolutionary run. We further show that the population moves significantly in genotype space during this neutral phase, possibly increasing in divergence. Finally, we investigate the probabilities of mutating to a higher fitness, above the neutral plateau, and find no evidence for a significant upward trend in these probabilities before the crucial mutations actually occurred.

[1]  Phil Husbands,et al.  Evolving Robot Behaviours with Diffusing Gas Networks , 1998, EvoRobot.

[2]  Weinberger,et al.  Local properties of Kauffman's N-k model: A tunably rugged energy landscape. , 1991, Physical review. A, Atomic, molecular, and optical physics.

[3]  Julian Francis Miller,et al.  Information Characteristics and the Structure of Landscapes , 2000, Evolutionary Computation.

[4]  M. Huynen,et al.  Smoothness within ruggedness: the role of neutrality in adaptation. , 1996, Proceedings of the National Academy of Sciences of the United States of America.

[5]  L. Barnett.  TANGLED WEBS Evolutionary Dynamics on Fitness Landscapes with Neutrality , 1997 .

[6]  Stuart A. Kauffman,et al.  The origins of order , 1993 .

[7]  P. Schuster,et al.  Analysis of RNA sequence structure maps by exhaustive enumeration I. Neutral networks , 1995 .

[8]  T. Jukes,et al.  The neutral theory of molecular evolution. , 2000, Genetics.

[9]  Terry Jones,et al.  Fitness Distance Correlation as a Measure of Problem Difficulty for Genetic Algorithms , 1995, ICGA.

[10]  M. Shackleton,et al.  An investigation of redundant genotype-phenotype mappings and their role in evolutionary search , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[11]  Christian M. Reidys,et al.  Evolutionary Dynamics and Optimization: Neutral Networks as Model-Landscapes for RNA Secondary-Structure Folding-Landscapes , 1995, ECAL.

[12]  Phil Husbands,et al.  Better Living Through Chemistry: Evolving GasNets for Robot Control , 1998, Connect. Sci..

[13]  M. Nowak,et al.  Error thresholds of replication in finite populations mutation frequencies and the onset of Muller's ratchet. , 1989, Journal of theoretical biology.

[14]  P. Husbands,et al.  Four-Dimensional Neuronal Signaling by Nitric Oxide: A Computational Analysis , 2000, The Journal of Neuroscience.

[15]  Julian Francis Miller,et al.  The Advantages of Landscape Neutrality in Digital Circuit Evolution , 2000, ICES.

[16]  Inman Harvey,et al.  Through the Labyrinth Evolution Finds a Way: A Silicon Ridge , 1996, ICES.

[17]  C. Hölscher.  Nitric oxide, the enigmatic neuronal messenger: its role in synaptic plasticity , 1997, Trends in Neurosciences.

[18]  Andrew Philippides,et al.  Nitric Oxide Signalling in Real and Artificial Neural Networks , 2000 .

[19]  M. Kimura.  The Neutral Theory of Molecular Evolution: Introduction , 1983 .

[20]  L. Barnett.  Ruggedness and neutrality—the NKp family of fitness landscapes , 1998 .