Hash-Based Tree Similarity and Simplification in Genetic Programming for Symbolic Regression

TL;DR

This paper presents a fast tree hashing algorithm for identifying isomorphic subtrees, enabling improved diversity management and algebraic simplification in genetic programming for symbolic regression, leading to promising benchmark results.

Contribution

The paper introduces a novel runtime-efficient tree hashing method for isomorphic subtrees and applies it to enhance diversity preservation and simplification in genetic programming.

Findings

Effective diversity preservation mechanism demonstrated

Improved symbolic regression performance on benchmarks

Fast online identification of isomorphic subtrees

Abstract

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.