Empirical analysis of different levels of meta-evolution

We analyze different levels of meta-evolution using a graph based GP system. The system allows one to represent individuals of the search space and genetic variation operators in a coherent way as graph programs differing only in the operator set. Seven variants of meta-evolution are tested on three real world classification problems. The most complex variant consists of three meta-levels where graph programs on meta-level 1 recombine individuals of the search space (base level), graph programs on meta-level 2 recombine programs on meta-level 1, and programs on meta-level 3 recombine programs on meta-level 2 and themselves. The empirical results shows that the use of meta levels is advantageous.

[1]  Astro Teller,et al.  PADO: a new learning architecture for object recognition , 1997 .

[2]  L. Darrell Whitley,et al.  Delta Coding: An Iterative Search Strategy for Genetic Algorithms , 1991, ICGA.

[3]  R. Rosenberg.  Simulation of genetic populations with biochemical properties : technical report , 1967 .

[4]  Annie S. Wu,et al.  Empirical Studies of the Genetic Algorithm with Noncoding Segments , 1995, Evolutionary Computation.

[5]  Lutz Prechelt,et al.  A Set of Neural Network Benchmark Problems and Benchmarking Rules , 1994 .

[6]  Wolfgang Banzhaf,et al.  Self-Evolution in a Constructive Binary String System , 1998, Artificial Life.

[7]  John R. Koza,et al.  Spontaneous Emergence of Self-Replicating and Evolutionarily Self-Improving Computer Programs , 1993 .

[8]  Robert I. Damper,et al.  VQ Score Normalisation for Text-dependent and Text-independent Speaker Recognition , 1997, AVBPA.

[9]  Hitoshi Iba,et al.  Extending genetic programming with recombinative guidance , 1996 .

[10]  Peter Nordin,et al.  A compiling genetic programming system that directly manipulates the machine-code , 1994 .

[11]  L. Darrell Whitley,et al.  Remapping Hyperspace During Genetic Search: Canonical Delta Folding , 1992, FOGA.

[12]  J. David Schaffer,et al.  An Adaptive Crossover Distribution Mechanism for Genetic Algorithms , 1987, ICGA.

[13]  Wolfgang Banzhaf,et al.  Hierarchical Genetic Programming using Local Modules , 2001 .

[14]  Hans-Paul Schwefel,et al.  Evolution and optimum seeking , 1995, Sixth-generation computer technology series.

[15]  Peter Nordin,et al.  Genetic programming - An Introduction: On the Automatic Evolution of Computer Programs and Its Applications , 1998 .

[16]  Riccardo Poli,et al.  Evolution of Graph-Like Programs with Parallel Distributed Genetic Programming , 1997, ICGA.

[17]  J. Pollack,et al.  Coevolving High-Level Representations , 1993 .

[18]  Thomas Bck,et al.  Self-adaptation in genetic algorithms , 1991 .

[19]  Wolfgang Banzhaf,et al.  Meta-Evolution in Graph GP , 1999, EuroGP.

[20]  David B. Fogel,et al.  Meta-evolutionary programming , 1991, [1991] Conference Record of the Twenty-Fifth Asilomar Conference on Signals, Systems & Computers.

[21]  Lawrence Davis,et al.  Adapting Operator Probabilities in Genetic Algorithms , 1989, ICGA.

[22]  Astro Teller,et al.  Evolving programmers: the co-evolution of intelligent recombination operators , 1996 .

[23]  C. G. Shaefer,et al.  The ARGOT Strategy: Adaptive Representation Genetic Optimizer Technique , 1987, ICGA.

[24]  Vidroha Debroy,et al.  Genetic Programming , 1998, Lecture Notes in Computer Science.

[25]  Justinian P. Rosca,et al.  Learning by adapting representations in genetic programming , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

[26]  Bruce Edmonds,et al.  Meta-Genetic Programming: Co-evolving the Operators of Variation , 2001 .

[27]  Peter J. Angeline,et al.  Two self-adaptive crossover operators for genetic programming , 1996 .