We used a meta-optimization environment to compare two reference versions of the Particle Swarm Optimization (PSO) algorithm, namely Standard PSO 2006 and Standard PSO 2011, on the CEC 2013 benchmark. We first compared the performances of the two standard PSO versions using the parameter sets suggested in the literature. Then, we automatically tuned the parameter values of both algorithms, to allow the two versions to perform at their best.
Automatic tuning of standard PSO versions / Ugolotti, Roberto; Cagnoni, Stefano. - ELETTRONICO. - (2015), pp. 1501-1502. (Intervento presentato al convegno 17th Genetic and Evolutionary Computation Conference, GECCO 2015 tenutosi a esp nel 2015) [10.1145/2739482.2764706].
Automatic tuning of standard PSO versions
Ugolotti, Roberto;Cagnoni, Stefano
2015-01-01
Abstract
We used a meta-optimization environment to compare two reference versions of the Particle Swarm Optimization (PSO) algorithm, namely Standard PSO 2006 and Standard PSO 2011, on the CEC 2013 benchmark. We first compared the performances of the two standard PSO versions using the parameter sets suggested in the literature. Then, we automatically tuned the parameter values of both algorithms, to allow the two versions to perform at their best.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.