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

Autor: Adrian Binding, Johannes Jakubik, Stefan Feuerriegel
Jazyk: angličtina
Rok vydání: 2021
Předmět:
FOS: Computer and information sciences
Mathematical optimization
Computer Science - Machine Learning
Information Systems and Management
General Computer Science
Computer science
0211 other engineering and technologies
Evolutionary algorithm
MathematicsofComputing_NUMERICALANALYSIS
Forecasting
Gaussian process
Surrogate model
SPSO2011
Particle swarm optimization
02 engineering and technology
Management Science and Operations Research
Industrial and Manufacturing Engineering
Machine Learning (cs.LG)
Set (abstract data type)
symbols.namesake
0502 economics and business
Neural and Evolutionary Computing (cs.NE)
050210 logistics & transportation
021103 operations research
05 social sciences
Bayesian optimization
Computer Science - Neural and Evolutionary Computing
Function (mathematics)
Modeling and Simulation
symbols
Benchmark (computing)
Zdroj: European Journal of Operational Research, 295 (1)
ISSN: 0377-2217
1872-6860
Popis: 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.
European Journal of Operational Research, 295 (1)
ISSN:0377-2217
ISSN:1872-6860
Databáze: OpenAIRE
Externí odkaz: