PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

When A Genetic Algorithm Outperforms Hill-Climbing
Adam Prügel-Bennett
Theoretical Computer Science Volume 320, Number 1, pp. 135-153, 2004.

Abstract

A toy optimisation problem is introduced which consists of a fitness gradient broken up by a series of hurdles. The performance of a hill-climber and a stochastic hill-climber are computed. These are compared with the empirically observed performance of a genetic algorithm (GA) with and without. The hill-climber with a sufficiently large neighbourhood outperforms the stochastic hill-climber, but is outperformed by a GA both with and without crossover. The GA with crossover substantially outperforms all the other heuristics considered here. The relevance of this result to real world problems is discussed.

Postscript - Requires a viewer, such as GhostView
EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:1770
Deposited By:Adam Prügel-Bennett
Deposited On:28 November 2005