Directed particle swarm optimization with Gaussian-process-based function forecasting

Open access
Date
2021-11-16Type
- Journal Article
Citations
Cited 8 times in
Web of Science
Cited 10 times in
Scopus
ETH Bibliography
yes
Altmetrics
Abstract
Particle swarm optimization (PSO) is an iterative search method that moves a set of candidate solution around a search-space towards the best known global and local solutions with randomized step lengths. PSO frequently accelerates optimization in practical applications, where gradients are not available and function evaluations expensive. Yet the traditional PSO algorithm ignores the potential knowledge that could have been gained of the objective function from the observations by individual particles. Hence, we draw upon concepts from Bayesian optimization and introduce a stochastic surrogate model of the objective function. That is, we fit a Gaussian process to past evaluations of the objective function, forecast its shape and then adapt the particle movements based on it. Our computational experiments demonstrate that baseline implementations of PSO (i. e., SPSO2011) are outperformed. Furthermore, compared to, state-of-art surrogate-assisted evolutionary algorithms, we achieve substantial performance improvements on several popular benchmark functions. Overall, we find that our algorithm attains desirable properties for exploratory and exploitative behavior. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000471306Publication status
publishedExternal links
Journal / series
European Journal of Operational ResearchVolume
Pages / Article No.
Publisher
ElsevierSubject
Forecasting; Gaussian process; Surrogate model; SPSO2011; Particle swarm optimizationOrganisational unit
09623 - Feuerriegel, Stefan (ehemalig) / Feuerriegel, Stefan (former)
More
Show all metadata
Citations
Cited 8 times in
Web of Science
Cited 10 times in
Scopus
ETH Bibliography
yes
Altmetrics