Publication detail

Combining Lipschitz and RBF Surrogate Models for High-dimensional Computationally Expensive Problems

KŮDELA, J. MATOUŠEK, R.

English title

Combining Lipschitz and RBF Surrogate Models for High-dimensional Computationally Expensive Problems

Type

journal article in Web of Science

Language

en

Original abstract

Standard evolutionary optimization algorithms assume that the evaluation of the objective and constraint functions is straightforward and computationally cheap. However, in many real-world optimization problems, these evaluations involve computationally expensive numerical simulations or physical experiments. Surrogate-assisted evolutionary algorithms (SAEAs) have recently gained increased attention for their performance in solving these types of problems. The main idea of SAEAs is the integration of an evolutionary algorithm with a selected surrogate model that approximates the computationally expensive function. In this paper, we propose a surrogate model based on a Lipschitz underestimation and use it to develop a differential evolution-based algorithm. The algorithm, called Lipschitz Surrogate-assisted Differential Evolution (LSADE), utilizes the Lipschitz-based surrogate model, along with a standard radial basis function surrogate model and a local search procedure. The experimental results on seven benchmark functions of dimensions 30, 50, 100, and 200 show that the proposed LSADE algorithm is competitive compared with the state-of-the-art algorithms under a limited computational budget, being especially effective for the very complicated benchmark functions in high dimensions.

English abstract

Standard evolutionary optimization algorithms assume that the evaluation of the objective and constraint functions is straightforward and computationally cheap. However, in many real-world optimization problems, these evaluations involve computationally expensive numerical simulations or physical experiments. Surrogate-assisted evolutionary algorithms (SAEAs) have recently gained increased attention for their performance in solving these types of problems. The main idea of SAEAs is the integration of an evolutionary algorithm with a selected surrogate model that approximates the computationally expensive function. In this paper, we propose a surrogate model based on a Lipschitz underestimation and use it to develop a differential evolution-based algorithm. The algorithm, called Lipschitz Surrogate-assisted Differential Evolution (LSADE), utilizes the Lipschitz-based surrogate model, along with a standard radial basis function surrogate model and a local search procedure. The experimental results on seven benchmark functions of dimensions 30, 50, 100, and 200 show that the proposed LSADE algorithm is competitive compared with the state-of-the-art algorithms under a limited computational budget, being especially effective for the very complicated benchmark functions in high dimensions.

Keywords in English

Lipschitz surrogate model; Differential evolution; Radial basis function; Surrogate assisted evolutionary algorithms; High-dimensional expensive optimization

Released

17.11.2022

ISSN

0020-0255

Volume

619

Number

January

Pages from–to

457–477

Pages count

21

BIBTEX


@article{BUT179504,
  author="Jakub {Kůdela} and Radomil {Matoušek},
  title="Combining Lipschitz and RBF Surrogate Models for High-dimensional Computationally Expensive Problems",
  year="2022",
  volume="619",
  number="January",
  month="November",
  pages="457--477",
  issn="0020-0255"
}