Tag-based Genetic Regulation for Genetic Programming

Abstract

We introduce and experimentally demonstrate tag-based genetic regulation, a new genetic programming (GP) technique that allows evolving programs to conditionally express code modules. Tags are evolvable names that provide a flexible mechanism for labeling and referring to code modules. Tag-based genetic regulation extends existing tag-based naming schemes to allow programs to promote and repress code modules. This extension allows evolution to structure a program as an arbitrary gene regulatory network where genes are program modules and program instructions mediate regulation. We demonstrate the functionality of tag-based regulation on several diagnostic tasks as well as a more challenging program synthesis problem. We find that tag-based regulation improves problem-solving performance on context-dependent problems where programs must adjust responses to particular inputs over time (e.g., based on local context). We also observe that our implementation of tag-based genetic regulation can impede adaptive evolution when expected outputs are not context-dependent (i.e., the correct response to a particular input remains static over time). Tag-based genetic regulation broadens our repertoire of techniques for evolving more dynamic genetic programs and can easily be incorporated into existing tag-enabled GP systems.