Hash-Based Tree Similarity and Simplification in Genetic Programming for Symbolic Regression
This work addresses efficiency and diversity issues in genetic programming for symbolic regression, though it appears incremental as it builds on existing hashing and simplification techniques.
The paper tackles the problem of efficiently identifying isomorphic subtrees in genetic programming for symbolic regression, resulting in a runtime-efficient tree hashing algorithm that enables fast population diversity calculation and algebraic simplification, with promising results on benchmark problems.
We introduce in this paper a runtime-efficient tree hashing algorithm for the identification of isomorphic subtrees, with two important applications in genetic programming for symbolic regression: fast, online calculation of population diversity and algebraic simplification of symbolic expression trees. Based on this hashing approach, we propose a simple diversity-preservation mechanism with promising results on a collection of symbolic regression benchmark problems.