Making and breaking power laws in evolutionary algorithm population dynamics

Deepening our understanding of the characteristics and behaviors of population-based search algorithms remains an important ongoing challenge in Evolutionary Computation. To date however, most studies of Evolutionary Algorithms have only been able to take place within tightly restricted experimental conditions. For instance, many analytical methods can only be applied to canonical algorithmic forms or can only evaluate evolution over simple test functions. Analysis of EA behavior under more complex conditions is needed to broaden our understanding of this population-based search process. This paper presents an approach to analyzing EA behavior that can be applied to a diverse range of algorithm designs and environmental conditions. The approach is based on evaluating an individual’s impact on population dynamics using metrics derived from genealogical graphs. From experiments conducted over a broad range of conditions, some important conclusions are drawn in this study. First, it is determined that very few individuals in an EA population have a significant influence on future population dynamics with the impact size fitting a power law distribution. The power law distribution indicates there is a non-negligible probability that single individuals will dominate the entire population, irrespective of population size. Two EA design features are however found to cause strong changes to this aspect of EA behavior: (1) the population topology and (2) the introduction of completely new individuals. If the EA population topology has a long path length or if new (i.e. historically uncoupled) individuals are continually inserted into the population, then power law deviations are observed for large impact sizes. It is concluded that such EA designs can not be dominated by a small number of individuals and hence should theoretically be capable of exhibiting higher degrees of parallel search behavior.

[1]  J. Crutchfield,et al.  Optimizing Epochal Evolutionary Search: Population-Size Dependent Theory , 1998, Machine Learning.

[2]  Zbigniew J. Czech,et al.  Selection Schemes in Evolutionary Algorithms , 2002, Intelligent Information Systems.

[3]  A. Bergman,et al.  Evolutionary capacitance as a general feature of complex gene networks , 2003, Nature.

[4]  Andreas Wagner,et al.  Neutralism and selectionism: a network-based reconciliation , 2008, Nature Reviews Genetics.

[5]  Kalyanmoy Deb,et al.  A Comparative Analysis of Selection Schemes Used in Genetic Algorithms , 1990, FOGA.

[6]  C. Schlichting.  Hidden Reaction Norms, Cryptic Genetic Variation, and Evolvability , 2008, Annals of the New York Academy of Sciences.

[7]  P. Husbands,et al.  Local evolvability of statistically neutral GasNet robot controllers. , 2003, Bio Systems.

[8]  Philip W. Hedrick,et al.  Genetic Hitchhiking: A New Factor in Evolution? , 1982 .

[9]  B. Garvik,et al.  Principles for the Buffering of Genetic Variation , 2001, Science.

[10]  G. Wagner,et al.  The Population Genetic Theory of Hidden Variation and Genetic Robustness , 2004, Genetics.

[11]  Nostrand Reinhold,et al.  the utility of using the genetic algorithm approach on the problem of Davis, L. (1991), Handbook of Genetic Algorithms. Van Nostrand Reinhold, New York. , 1991 .

[12]  Lawrence. Davis,et al.  Handbook Of Genetic Algorithms , 1990 .

[13]  Alden H. Wright,et al.  Genetic Algorithms for Real Parameter Optimization , 1990, FOGA.

[14]  W Stephan,et al.  Linkage Disequilibrium Under Genetic Hitchhiking in Finite Populations , 2008, Genetics.

[15]  Bryant A. Julstrom,et al.  What Have You Done for Me Lately? Adapting Operator Probabilities in a Steady-State Genetic Algorithm , 1995, ICGA.

[16]  Samir W. Mahfoud.  A Comparison of Parallel and Sequential Niching Methods , 1995, ICGA.

[17]  Greg Gibson,et al.  Uncovering cryptic genetic variation , 2004, Nature Reviews Genetics.

[18]  Bryant A. Julstrom,et al.  Adaptive operator probabilities in a genetic algorithm that applies three operators , 1997, SAC '97.

[19]  Thomas Bäck,et al.  An analysis of the behavior of simplified evolutionary algorithms on trap functions , 2003, IEEE Trans. Evol. Comput..

[20]  Tobias Blickle,et al.  Theory of evolutionary algorithms and application to system synthesis , 1997 .

[21]  Ruhul A. Sarker,et al.  Credit assignment in adaptive evolutionary algorithms , 2006, GECCO '06.

[22]  James M. Whitacre,et al.  Adaptation and self-organization in evolutionary algorithms , 2009, ArXiv.

[23]  A. Wagner,et al.  Robustness and evolution: concepts, insights and challenges from a developmental model system , 2008, Heredity.