ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Binomially distributed populations for modelling GAs

Srinivas, M and Patnaik, LM (1993) Binomially distributed populations for modelling GAs. In: Proceedings of the 5th International Conference on Genetic Algorithms, 17-22 July 1993, San Francisco, CA, USA, pp. 138-145.

Full text not available from this repository. (Request a copy)
Official URL: http://dl.acm.org/citation.cfm?id=657763&dl=GUIDE&...


We discuss a novel model for analyzing the working of genetic algorithms, when the objective function is a function of unitation. The model is exact (not approximate), and is valid for infinite populations. We introduce the notion of a binomially distributed population (BDP) as the building block of our model, and we show that the effect of uniform crossover on BDPs is to generate two other BDPs. We demonstrate that a population with any general distribution may be decomposed into several BDPs. We also show that a general multipoint crossover may be considered as a composition of several uniform crossovers. Based on these results, the effects of mutation and crossover on the distribution of strings have been characterized, and the model has been defined. The time complexity of the algorithm derived from the model is $O(\int^3)$ (where $\int$ is the problem size), a significant improvement over previous models with exponential time complexities, GASIM-a Genetic Algorithm Simulator for functions of unitation-has been implemented, and the exactness of the results obtained from GASIM has been verified from actual genetic algorithm runs

Item Type: Conference Paper
Publisher: Morgan Kaufmann Publishers Inc.
Additional Information: Copyright of this article belongs to Morgan Kaufmann Publishers Inc.
Keywords: computational complexity;genetic algorithms
Department/Centre: Division of Electrical Sciences > Computer Science & Automation
Date Deposited: 09 Jan 2008
Last Modified: 12 Jan 2012 06:43
URI: http://eprints.iisc.ac.in/id/eprint/11103

Actions (login required)

View Item View Item