Surrogate-Assisted Evolutionary Optimization of Large Problems
Chugh, T., Sun, C., Wang, H., & Jin, Y. (2020). Surrogate-Assisted Evolutionary Optimization of Large Problems. In T. Bartz-Beielstein, B. Filipič, P. Korošec, & E.-G. Talbi (Eds.), High-Performance Simulation-Based Optimization (pp. 165-187). Studies in Computational Intelligence, 833. Cham: Springer. doi:10.1007/978-3-030-18764-4_8
Published inStudies in Computational Intelligence
© Springer Nature Switzerland AG 2020.
This chapter presents some recent advances in surrogate-assisted evolutionary optimization of large problems. By large problems, we mean either the number of decision variables is large, or the number of objectives is large, or both. These problems pose challenges to evolutionary algorithms themselves, constructing surrogates and surrogate management. To address these challenges, we proposed two algorithms, one called kriging-assisted reference vector guided evolutionary algorithm (K-RVEA) for many-objective optimization, and the other called cooperative swarm optimization algorithm (SA-COSO) for high-dimensional single-objective optimization. Empirical studies demonstrate that K-RVEA works well for many-objective problems having up to ten objectives, while SA-COSA outperforms the state-of-the-art algorithms on 200-dimensional single-objective test problems.