Parameterless evolutionary algorithm applied to the nuclear reload problem

Autor:	Gustavo Henrique Flores Caldas, Roberto Schirru
Rok vydání:	2008
Předmět:	education.field_of_study Nuclear Energy and Engineering Computer science business.industry Population-based incremental learning Population Incremental learning Evolutionary algorithm Artificial intelligence education business
Zdroj:	Annals of Nuclear Energy. 35:583-590
ISSN:	0306-4549
DOI:	10.1016/j.anucene.2007.08.014
Popis:	In this work, an evolutionary algorithm with no parameters called FPBIL (parameter free PBIL) is developed based on PBIL (population-based incremental learning). Moreover, the analysis reveals how the parameters from PBIL can be replaced by self-adaptable mechanisms which appear from the radically different form by which the evolution is processed. Despite the advantages, the FPBIL reveals itself compact and relatively modest in the use of computational resources. The FPBIL is then applied to the nuclear reload problem. The experimental results observed are compared to those of other works and corroborate to affirm the superiority of the new algorithm.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::dac04547ff16fb671d03b1eb6dc99108 https://doi.org/10.1016/j.anucene.2007.08.014 Zobrazit plný text záznamu Full Text from ScienceDirect