This article describes model-based methods for global optimization. After introducing the global optimization framework, modeling approaches for stochastic algorithms are presented. We differentiate between models that use a distribution and models that use an explicit surrogate model. Fundamental aspects of and recent advances in surrogate-model based optimization are discussed. Strategies for selecting and evaluating surrogates are presented. The article concludes with a description of key features of two state-of-the-art surrogate model based algorithms, namely the evolvability learning of surrogates (EvoLS) algorithm and the sequential parameter optimization (SPO).

This lecture is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 692286.