Abstract
It becomes increasingly difficult to train a high-quality surrogate model as the dimension of a problem increases, especially for expensive optimization problems where only a limited number of samples can be afforded. This chapter focuses on addressing high-dimensional expensive problems that have over 30 and up to some 200 decision variables. The main techniques include the use of more exploratory search, co-operative search between multiple populations assisted by multiple surrogates, introduction of a Pareto-based infill criterion, decomposition of the decision space by random subsampling, and multi-tasking evolutionary optimization.