Adapting Representation in Genetic Programming

Genetic Programming uses trees to represent chromosomes. The user defines the representation space by defining the set of functions and terminals to label the nodes in the trees. The sufficiency principle requires that the set be sufficient to label the desired solution trees. To satisfy this principle, the user is often forced to provide a large set, which unfortunately also enlarges the representation space and thus, the search space. Structure-preserving crossover, STGP, CGP, and CFG-based GP, give the user the power to reduce the space by specifying rules for valid tree construction. However, the user often may not be aware of the best representation space, including heuristics, to solve a particular problem. In this paper, we present a methodology, which extracts and utilizes local heuristics aiming at improving search efficiency. The methodology uses a specific technique for extracting the heuristics, based on tracing firstorder (parent-child) distributions of functions and t...

• Type: Conference
• Year: 2004
• Where: GECCO
• Authors: Cezary Z. Janikow
• Keywords: GECCO 2004, Programming Uses Trees, Representation Space, User