Self-Adaptive Surrogate-Assisted Covariance Matrix Adaptation Evolution Strategy - Archive ouverte HAL Access content directly
Conference Papers Year : 2012

Self-Adaptive Surrogate-Assisted Covariance Matrix Adaptation Evolution Strategy

(1) , (1, 2) , (1, 3)
1
2
3

Abstract

This paper presents a novel mechanism to adapt surrogate-assisted population-based algorithms. This mechanism is applied to ACM-ES, a recently proposed surrogate-assisted variant of CMA-ES. The resulting algorithm, saACM-ES, adjusts online the lifelength of the current surrogate model (the number of CMA-ES generations before learning a new surrogate) and the surrogate hyper-parameters. Both heuristics significantly improve the quality of the surrogate model, yielding a significant speed-up of saACM-ES compared to the ACM-ES and CMA-ES baselines. The empirical validation of saACM-ES on the BBOB-2012 noiseless testbed demonstrates the efficiency and the scalability w.r.t the problem dimension and the population size of the proposed approach, that reaches new best results on some of the benchmark problems.
Fichier principal
Vignette du fichier
paper511.pdf (395.67 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00686570 , version 1 (10-04-2012)

Identifiers

Cite

Ilya Loshchilov, Marc Schoenauer, Michèle Sebag. Self-Adaptive Surrogate-Assisted Covariance Matrix Adaptation Evolution Strategy. Genetic and Evolutionary Computation Conference (GECCO 2012), Jul 2012, Philadelphia, United States. pp.321-328. ⟨hal-00686570⟩
184 View
376 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More