The goal of the paper is to provide an overview of classical and agent-based models of parallel evolutionary algorithms. Agent approach reveals possibilities of unification of various models and thus allows for the development of platforms supporting the implementation of different PEA variants. Design considerations based on AgWorld and Ant.NET projects conclude the paper. © Springer-Verlag 2004.
CITATION STYLE
Kisiel-Dorohinicki, M. (2004). Agent-based models and platforms for parallel evolutionary algorithms. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3038, 646–653. https://doi.org/10.1007/978-3-540-24688-6_84
Mendeley helps you to discover research relevant for your work.