A surrogate model assisted evolutionary algorithm for computationally expensive design optimization problems with discrete variables

Abstract

Abstract-Real-world computationally expensive design optimization problems with discrete variables pose challenges to surrogate-based optimization methods in terms of both efficiency and search ability. In this paper, a new method is introduced, called surrogate model-aware differential evolution with neighbourhood exploration, which has two phases. The first phase adopts a surrogate-based optimization method based on efficient surrogate model-aware search framework, the goal of which is to reach at least the neighbourhood of the global optimum. In the second phase, a neighbourhood exploration method for discrete variables is developed and collaborates with the first phase to further improve the obtained solutions. Empirical studies on various benchmark problems and a realworld network-on-chip design optimization problem show the combined advantages in terms of efficiency and search ability: when only a very limited number of exact evaluations are allowed, the proposed method is not slower than one of the most efficient methods for the targeted problem; when more evaluations are allowed, the proposed method can obtain results with comparable quality compared to standard differential evolution, but it requires only 1% to 30% of exact function evaluations.

Rule‐based searching for collision test cases of autonomous vehicles simulation

“…The method consists of two simulation models [19] and one main searching algorithm and its three sub-algorithms.…”

Restricted Boltzmann Machine‐Assisted Estimation of Distribution Algorithm for Complex Problems

“…To further verify the performance of the RBMAEDA, RBMAEDA is compared with two popular SAEAs, the SMAS algorithm [18] and the SMDN algorithm [19] presented in 2016, on a series of benchmark problems with a limited computation budget. In particular, the SMDN algorithm is a state-of-the-art algorithm for complex computationally expensive optimization problems with discrete variables, using the same comparison function to compare with the results provided in [19]. In the experiment, three indicators are used to measure the performance of these algorithms: 9 Complexity the average value, the standard deviation, and the success rate in the 10 trials.…”

“…In solving such computationally expensive optimization problems, the heavy computational cost has a major impact on the effectiveness and efficiency of traditional ECs. To address this challenge, surrogate-assisted evolutionary algorithms (SAEAs) [10][11][12], such as surrogate-assisted GA [13,14], surrogate-assisted PSO [15][16][17], and surrogateassisted DE [18,19], are receiving increasing attention in the EC community.…”

“…However, the SMAS are trapped in local optima for computationally expensive optimization problems with discrete variables. Furthermore, Liu et al [19] proposed a SAEA to solve expensive design optimization problems with discrete variables. Chugh et al [25] proposed a krigingbased surrogate-assisted reference vector-guided evolutionary algorithm to approximate the computationally expensive objective function to reduce the computational cost.…”