A distance between populations for one-point crossover in genetic algorithms

Genetic algorithms use transformation operators on the genotypic structures of the individuals to carry out a search. These operators define a neighborhood. To analyze various dynamics of the search process, it is often useful to define a distance in this space. In fact, using an operator-based distance can make the analysis more accurate and reliable than using distances which have no relationship with the genetic operators. In this paper we define a distance which is based on the standard one-point crossover. Given that the population strongly affects the neighborhood induced by the crossover, we first define a crossover-based distance between populations. Successively, we show that it is naturally possible to derive from this function a family of distances between individuals. Finally, we also introduce an algorithm to compute this distance efficiently.

[1]  Alden H. Wright,et al.  Representation Invariant Genetic Operators , 2010, Evolutionary Computation.

[2]  Leonardo Vanneschi,et al.  Diversity in Multipopulation Genetic Programming , 2003, GECCO.

[3]  Riccardo Poli,et al.  Topological Interpretation of Crossover , 2004, GECCO.

[4]  Alberto Moraglio,et al.  One-Point Geometric Crossover , 2010, PPSN.

[5]  Leonardo Vanneschi,et al.  Theory and practice for efficient genetic programming , 2004 .

[6]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[7]  Peter F. Stadler,et al.  Algebraic Theory of Recombination Spaces , 1997, Evolutionary Computation.

[8]  Graham Kendall,et al.  Problem Difficulty and Code Growth in Genetic Programming , 2004, Genetic Programming and Evolvable Machines.

[9]  Nikolay I. Nikolaev,et al.  Fitness Landscapes and Inductive Genetic Programming , 1997, ICANNGA.

[10]  Leonardo Vanneschi,et al.  A Study of Diversity in Multipopulation Genetic Programming , 2003, Artificial Evolution.

[11]  Thomas Jansen,et al.  06061 Executive Summary -- Theory of Evolutionary Algoritms , 2006, Theory of Evolutionary Algorithms.

[12]  Graham Kendall,et al.  Diversity in genetic programming: an analysis of measures and correlation with fitness , 2004, IEEE Transactions on Evolutionary Computation.

[13]  Peter F. Stadler,et al.  Recombination Spaces, Metrics, and Pretopologies , 2002 .

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

[15]  Alberto Moraglio,et al.  Geometry of evolutionary algorithms , 2011, GECCO.

[16]  Peter F. Stadler,et al.  Complex Adaptations and the Structure of Recombination Spaces , 1997 .

[17]  D. E. Goldberg,et al.  Genetic Algorithms in Search , 1989 .

[18]  H. Poincaré.  La science et l'hypothèse , 1968 .

[19]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[20]  Michael D. Vose,et al.  The simple genetic algorithm - foundations and theory , 1999, Complex adaptive systems.

[21]  Alden H. Wright,et al.  Neighborhood graphs and symmetric genetic operators , 2007, FOGA'07.

[22]  Anne Auger,et al.  08051 Executive Summary - Theory of Evolutionary Algorithms , 2008, Theory of Evolutionary Algorithms.

[23]  R. Rajendiran,et al.  Topological Spaces , 2019, A Physicist's Introduction to Algebraic Structures.

[24]  Stefan Friedrich,et al.  Topology , 2019, Arch. Formal Proofs.

[25]  Colin R. Reeves,et al.  Genetic Algorithms: Principles and Perspectives: A Guide to Ga Theory , 2002 .

[26]  Leonardo Vanneschi,et al.  How Far Is It from Here to There? A Distance That Is Coherent with GP Operators , 2011, EuroGP.

[27]  Leonardo Vanneschi,et al.  A Study of Fitness Distance Correlation as a Difficulty Measure in Genetic Programming , 2005, Evolutionary Computation.

[28]  Anikó Ekárt,et al.  Maintaining the Diversity of Genetic Programs , 2002, EuroGP.

[29]  Melanie Mitchell,et al.  The royal road for genetic algorithms: Fitness landscapes and GA performance , 1991 .

[30]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[31]  Victor J. Rayward-Smith,et al.  Fitness Distance Correlation and Ridge Functions , 1998, PPSN.

[32]  Marco Tomassini,et al.  Evolutionary Algorithms , 1995, Towards Evolvable Hardware.

[33]  Michael D. Vose.  Course notes: genetic algorithm theory , 2010, GECCO '10.

[34]  Kalyanmoy Deb,et al.  Analyzing Deception in Trap Functions , 1992, FOGA.