Netcrawling-optimal evolutionary search with neutral networks

Several studies have demonstrated that in the presence of a high degree of selective neutrality, in particular on fitness landscapes featuring neutral networks, evolution is qualitatively different from that of the more common model of rugged/correlated fitness landscapes often (implicitly) assumed by GA researchers. We characterise evolutionary dynamics on fitness landscapes with neutral networks and argue that, if a certain correlation-like statistical property holds, the most efficient strategy for evolutionary search is not population-based, but rather a population-of-one netcrawler-a variety of hill-climber. We derive quantitative estimates for expected waiting times to discovery of fitter genotypes and discuss implications for evolutionary algorithm design, including a proposal for an adaptive variant of the netcrawler.

[1]  Samuel Karlin,et al.  A First Course on Stochastic Processes , 1968 .

[2]  J. Kingman.  A FIRST COURSE IN STOCHASTIC PROCESSES , 1967 .

[3]  M. Kimura,et al.  An introduction to population genetics theory , 1971 .

[4]  P. Feldman.  Evolution of sex , 1975, Nature.

[5]  A. Kondrashov.  Selection against harmful mutations in large sexual and asexual populations. , 1982, Genetical research.

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

[7]  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.

[8]  Thomas Bäck,et al.  A Survey of Evolution Strategies , 1991, ICGA.

[9]  B. Derrida,et al.  Evolution in a flat fitness landscape , 1991 .

[10]  Melanie Mitchell,et al.  Relative Building-Block Fitness and the Building Block Hypothesis , 1992, FOGA.

[11]  Inman Harvey,et al.  Evolving visually guided robots , 1993 .

[12]  Weinberger,et al.  RNA folding and combinatory landscapes. , 1993, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[13]  Inman Harvey,et al.  Evolutionary robotics and SAGA: The case for hill crawling and tournament selection , 1994 .

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

[15]  D. Wolpert,et al.  No Free Lunch Theorems for Search , 1995 .

[16]  Inman Harvey,et al.  Noise and the Reality Gap: The Use of Simulation in Evolutionary Robotics , 1995, ECAL.

[17]  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.

[18]  Adrian Thompson,et al.  Silicon evolution , 1996 .

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

[20]  P. Schuster,et al.  Generic properties of combinatory maps: neutral networks of RNA secondary structures. , 1997, Bulletin of mathematical biology.

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

[22]  Matthew Quinn,et al.  Some Problems (and a Few Solutions) for Open-Ended Evolutionary Robotics , 1998, EvoRobots.

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

[24]  James P. Crutchfield,et al.  Statistical Dynamics of the Royal Road Genetic Algorithm , 1999, Theor. Comput. Sci..

[25]  Paul J. Layzell,et al.  Explorations in design space: unconventional electronics design through artificial evolution , 1999, IEEE Trans. Evol. Comput..

[26]  J. Crutchfield,et al.  Metastable evolutionary dynamics: Crossing fitness barriers or escaping via neutral paths? , 1999, Bulletin of mathematical biology.

[27]  Paul J. Layzell,et al.  Evolution of Robustness in an Electronics Design , 2000, ICES.

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

[29]  Christian M. Reidys,et al.  Neutrality in fitness landscapes , 2001, Appl. Math. Comput..