On the Workings of Genetic Algorithms: The Genoclique Fixing Hypothesis - Computer Science > Neural and Evolutionary ComputingReport as inadecuate




On the Workings of Genetic Algorithms: The Genoclique Fixing Hypothesis - Computer Science > Neural and Evolutionary Computing - Download this document for free, or read online. Document in PDF available to download.

Abstract: We recently reported that the simple genetic algorithm SGA is capable ofperforming a remarkable form of sublinear computation which has astraightforward connection with the general problem of interacting attributesin data-mining. In this paper we explain how the SGA can leverage thiscomputational proficiency to perform efficient adaptation on a broad class offitness functions. Based on the relative ease with which a practical fitnessfunction might belong to this broad class, we submit a new hypothesis about theworkings of genetic algorithms. We explain why our hypothesis is superior tothe building block hypothesis, and, by way of empirical validation, we presentthe results of an experiment in which the use of a simple mechanism calledclamping dramatically improved the performance of an SGA with uniform crossoveron large, randomly generated instances of the MAX 3-SAT problem.



Author: Keki M. Burjorjee

Source: https://arxiv.org/







Related documents