Improving Evaluation of Recombination-based Cartesian Genetic Programming
For researchers in genetic programming, this work demonstrates that recombination-based CGP can be made competitive through hyperparameter tuning, addressing a known limitation.
This study shows that hyperparameter optimization improves the performance of recombination-based Cartesian Genetic Programming on symbolic regression benchmarks, where subgraph crossover and discrete phenotypic recombination operators were tested on the SRBench platform.
Cartesian Genetic Programming has traditionally been using mutation as its main and often sole genetic operator to drive evolutionary search. Despite advancements in recent years, recombinationbased approaches have long been avoided, due to apparent lack of performance gains. This study examines two recently suggested recombination-based operators, subgraph crossover and discrete phenotypic recombination on SRBench, a benchmarking platform for symbolic regression. Using the implementations provided in the TinyverseGP framework, we perform hyperparameter optimisation of the respective representations with these two operators. Our work demonstrates that hyperparameter optimisation can lead to improvements in performance for recombination-based Cartesian Genetic Programming.