The concept of information gain has been adopted as tool to study the effectiveness of population-based optimizers; using this approach, it is possible to infer convergence properties for population-based optimizers. The experimental results have shown characteristic phase transition between exploration and exploitation phase during the evolutionary process and, moreover, the evidence that gain maximization offers a robust theoretical framework to study the convergence of stochastic optimizers.
Entropic Divergence for Population Based Optimization Algorithms
CUTELLO, Vincenzo;NICOSIA, GIUSEPPE;PAVONE, MARIO FRANCESCO;
2010-01-01
Abstract
The concept of information gain has been adopted as tool to study the effectiveness of population-based optimizers; using this approach, it is possible to infer convergence properties for population-based optimizers. The experimental results have shown characteristic phase transition between exploration and exploitation phase during the evolutionary process and, moreover, the evidence that gain maximization offers a robust theoretical framework to study the convergence of stochastic optimizers.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.