Optimization with surrogate models embedded

Data-driven surrogate models can learn nonlinear input-output relations and replace expensive simulations or experiments in optimization studies.

ReLU artificial neural networks (ANNs) can be used to approximate complex functions from data. In order to embed these functions into optimization problems, strong network formulations are needed. We develop methods and software that employ progressive bound tightening procedures to produce MIP encodings for ReLU networks. This allows users to embed complex and nonlinear functions into mixed-integer programs.

Key publications

• Schweidtmann, A. M., Esche, E., Fischer, A., Kloft, M., Repke, J. U., Sager, S., & Mitsos, A. (2021). Machine Learning in Chemical Engineering: A Perspective. Chemie Ingenieur Technik.

• Deterministic global optimization with Gaussian processes embedded [Schweidtmann et al., 2021]

• Deterministic global optimization with artificial neural networks embedded [Schweidtmann & Mitsos, 2019]