Differential Evolution (DE) is a popular and efficient continuous optimization technique based on the principles of Darwinian evolution. In a previous work we have introduced Asynchronous Differential Evolution (ADE), a DE generalization that allows to regulate the synchronization mechanism of the algorithm by tuning two additional parameters. However, it is well known that an evolutionary algorithm must have a small number of parameters in order to be easy to use. Moving from this consideration, in this paper we introduce two different selfadaptive ADE schemes built on top of jDE-2, one of the most efficient self-adaptive DE variant. Experiments on well known benchmark functions have been held and the performances of the proposed SaADE schemes are compared with those of jDE- 2.

Self-adaptive Asynchronous Differential Evolution

Santucci Valentino;
2011

Abstract

Differential Evolution (DE) is a popular and efficient continuous optimization technique based on the principles of Darwinian evolution. In a previous work we have introduced Asynchronous Differential Evolution (ADE), a DE generalization that allows to regulate the synchronization mechanism of the algorithm by tuning two additional parameters. However, it is well known that an evolutionary algorithm must have a small number of parameters in order to be easy to use. Moving from this consideration, in this paper we introduce two different selfadaptive ADE schemes built on top of jDE-2, one of the most efficient self-adaptive DE variant. Experiments on well known benchmark functions have been held and the performances of the proposed SaADE schemes are compared with those of jDE- 2.
9781612840727
File in questo prodotto:
File Dimensione Formato  
ssci2011_santucci_milani.pdf

non disponibili

Tipologia: Documento in Pre-print
Licenza: Creative commons
Dimensione 231.4 kB
Formato Adobe PDF
231.4 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/20.500.12071/11240
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact