Adaptive representation for single objective optimization

Authors:
Crina Grosan;Mihai Oltean
Affiliations:
Department of Computer Science, Faculty of Mathematics and Computer Science, Babes-Bolyai University, Kogalniceanu 1, 3400, Cluj-Napoca, Romania;Department of Computer Science, Faculty of Mathematics and Computer Science, Babes-Bolyai University, Kogalniceanu 1, 3400, Cluj-Napoca, Romania
Venue:
Soft Computing - A Fusion of Foundations, Methodologies and Applications
Year:
2005

Citing 0
Cited 3

Multicriteria programming in medical diagnosis and treatments

Applied Soft Computing
An Adaptive GP Strategy for Evolving Digital Circuits

KES '08 Proceedings of the 12th international conference on Knowledge-Based Intelligent Information and Engineering Systems, Part III
An autonomous GP-based system for regression and classification problems

Applied Soft Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new technique called Adaptive Representation Evolutionary Algorithm (AREA) is proposed in this paper. AREA involves dynamic alphabets for encoding solutions. The proposed adaptive representation is more compact than binary representation. Genetic operators are usually more aggressive when higher alphabets are used. Therefore the proposed encoding ensures an efficient exploration of the search space. This technique may be used for single and multiobjective optimization. We treat the case of single objective optimization problems in this paper. Despite its simplicity the AREA method is able to generate a population converging towards optimal solutions. Numerical experiments indicate that the AREA technique performs better than other single objective evolutionary algorithms on the considered test functions.