Abstract

Types have been introduced to Genetic Programming (GP) by researchers with different motivation. We present the concept of types in GP and introduce a typed GP system, PolyGP, that supports polymorphism through the use of three different kinds of type variable. We demonstrate the usefulness of this kind of polymorphism in GP by evolving two polymorphic programs (nth and map) using the system. Based on the analysis of a series of experimental results, we conclude that this implementation of polymorphism is effective in assisting GP evolutionary search to generate these two programs. PolyGP may enhance the applicability of GP to a new class of problems that are difficult for other polymorphic GP systems to solve.

Citations

2458	Genetic Programming: On the Programming of Computers by Means of Natural Selection - Koza - 1992
727	On understanding types, data abstraction, and polymorphism - Cardelli, Wegner - 1985
206	Strongly typed genetic programming - Montana - 1993
62	Strongly typed genetic programming in evolving cooperation strategies - Haynes, Wainwright, et al. - 1995
45	Type inheritance in strongly typed genetic programming - Haynes, Schoenefeld, et al. - 1996
26	Evolving recursive programs for tree search - Brave - 1996
23	Methods to evolve legal phenotypes - Yu, Bentley - 1998
10	PolyGP: A polymorphic genetic programming system in haskell - Yu, Clack - 1998
7	Scaling of program fitness spaces - Langdon - 1999
6	private communication - Luke - 2008
5	Impact of types on essentially typeless problems in GP - McPhee, Hopper, et al. - 1998
4	An Analysis of the Impact of - Yu - 1999
2	An adaptive inductive logic programming system using genetic programming. Evolutionary Programming - Wong, Leung - 1995