Using Dissortative Mating Genetic Algorithms to Track the Extrema of Dynamic Deceptive Functions - Computer Science > Neural and Evolutionary ComputingReport as inadecuate




Using Dissortative Mating Genetic Algorithms to Track the Extrema of Dynamic Deceptive Functions - Computer Science > Neural and Evolutionary Computing - Download this document for free, or read online. Document in PDF available to download.

Abstract: Traditional Genetic Algorithms GAs mating schemes select individuals forcrossover independently of their genotypic or phenotypic similarities. InNature, this behaviour is known as random mating. However, non-random schemes -in which individuals mate according to their kinship or likeness - are morecommon in natural systems. Previous studies indicate that, when applied to GAs,negative assortative mating a specific type of non-random mating, also knownas dissortative mating may improve their performance on both speed andreliability in a wide range of problems. Dissortative mating maintains thegenetic diversity at a higher level during the run, and that fact is frequentlyobserved as an explanation for dissortative GAs ability to escape local optimatraps. Dynamic problems, due to their specificities, demand special care whentuning a GA, because diversity plays an even more crucial role than it doeswhen tackling static ones. This paper investigates the behaviour ofdissortative mating GAs, namely the recently proposed Adaptive DissortativeMating GA ADMGA, on dynamic trap functions. ADMGA selects parents accordingto their Hamming distance, via a self-adjustable threshold value. The method,by keeping population diversity during the run, provides an effective means todeal with dynamic problems. Tests conducted with deceptive and nearly deceptivetrap functions indicate that ADMGA is able to outperform other GAs, somespecifically designed for tracking moving extrema, on a wide range of tests,being particularly effective when speed of change is not very fast. Whencomparing the algorithm to a previously proposed dissortative GA, results showthat performance is equivalent on the majority of the experiments, but ADMGAperforms better when solving the hardest instances of the test set.



Author: C. M. Fernandes, J.J. Merelo, A.C. Rosa

Source: https://arxiv.org/



DOWNLOAD PDF




Related documents