Shortcomings with Tree-Structured Edge Encodings for Neural Networks

In evolutionary algorithms a common method for encoding neural networks is to use a tree-structured assembly procedure for constructing them. Since node operators have difficulties in specifying edge weights and these operators are execution-order dependent, an alternative is to use edge operators. Here we identify three problems with edge operators: in the initialization phase most randomly created genotypes produce an incorrect number of inputs and outputs; variation operators can easily change the number of input/output (I/O) units; and units have a connectivity bias based on their order of creation. Instead of creating I/O nodes as part of the construction process we propose using parameterized operators to connect to pre-existing I/O units. Results from experiments show that these parameterized operators greatly improve the probability of creating and maintaining networks with the correct number of I/O units, remove the connectivity bias with I/O units and produce better controllers for a goal-scoring task.

[1]  John R. Koza,et al.  Genetic programming - on the programming of computers by means of natural selection , 1993, Complex adaptive systems.

[2]  John R. Koza,et al.  Genetic Programming III: Darwinian Invention & Problem Solving , 1999 .

[3]  Riccardo Poli,et al.  Schema Theory for Genetic Programming with One-Point Crossover and Point Mutation , 1997, Evolutionary Computation.

[4]  P. Nordin.  Genetic Programming III - Darwinian Invention and Problem Solving , 1999 .

[5]  Gregory S. Hornby,et al.  Body-brain co-evolution using L-systems as a generative encoding , 2001 .

[6]  Phil Husbands,et al.  Evolutionary robotics , 2014, Evolutionary Intelligence.

[7]  Masahiro Fujita,et al.  Evolution of Controllers from a High-Level Simulator to a High DOF Robot , 2000, ICES.

[8]  Randall D. Beer,et al.  Evolving Dynamical Neural Networks for Adaptive Behavior , 1992, Adapt. Behav..

[9]  Julian F. Miller,et al.  Genetic and Evolutionary Computation — GECCO 2003 , 2003, Lecture Notes in Computer Science.

[10]  B. Mirtich,et al.  Diffuse versus true coevolution in a physics-based world , 1999 .

[11]  Lee Spector,et al.  Evolving Graphs and Networks with Edge Encoding: Preliminary Report , 1996 .

[12]  Scott Brave,et al.  Evolving deterministic finite automata using cellular encoding , 1996 .

[13]  Huebner,et al.  Proceedings of the First Annual Conference of the Wharton School of Finance and Commerce , 2022 .

[14]  Jordan B. Pollack,et al.  Creating High-Level Components with a Generative Representation for Body-Brain Evolution , 2002, Artificial Life.

[15]  下平 丕作士,et al.  The Genetic and Evolutionary Computation Conference , 2002 .