Regulatory genotype-to-phenotype mappings improve evolvability in genetic programming
Jin‐Ting Zhang, Ting Hu · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2022
Most genotype-to-phenotype mappings in EAs are redundant, i.e., multiple genotypes can map to the same phenotype. Phenotypes are accessible from one to another through point mutations. However, these mutational connections can be unevenly distributed among phenotypes. Quantitative analysis of such connections helps better characterize the robustness and evolvability of an EA. In this study, we propose two genotype-to-phenotype mapping mechanisms for linear genetic programming (LGP), where the execution and output of a linear genetic program are varied by a regulator. We investigate how such regulatory mappings can alter the genotypic connections among different phenotypes and the robustness and evolvability of phenotypes. We also compare the search ability of LGP using the conventional mapping versus the regulatory mappings, and observe that the regulatory mappings improve the efficiency in all three search scenarios, including random walk, hill climbing, and novelty search.