The Evolution of Genetic Algorithms: Towards Massive Parallelism

June 1993

The Evolution of Genetic Algorithms: Towards Massive Parallelism

Authors:

Shumeet Baluja
Conference Paper
Proceedings of (ICML) International Conference on Machine Learning

Abstract:

One of the issues in creating any search technique is balancing the need for diverse exploration with the desire for efficient focusing. This paper explores a genetic algorithm (GA) architecture which is more resilient to local optima than other recently introduced GA models. and which provides the ability to focus search quickly. The GA uses a fine-grain parallel architecture to simulate evolution more closely than previous models. In order to motivate the need for fine-grain parallelism, this paper will provide an overview of the two preceding phases of development: the traditional genetic algorithm, and the coarse-grain parallel GA. A test set of 15 problems is used to compare the effectiveness of a fine-grain parallel GA with that of a coarse-grain parallel GA.
@conference{Baluja-1993-15930,
author = {Shumeet Baluja},
title = {The Evolution of Genetic Algorithms: Towards Massive Parallelism},
booktitle = {Proceedings of (ICML) International Conference on Machine Learning},
year = {1993},
month = {June},
editor = {P.E. Utgoff},
pages = {1 - 8},
publisher = {Morgan Kaufmann Publishers},
}
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.