Constraint Learning to Define Trust Regions in Optimization over Pre-Trained Predictive Models

There is a recent proliferation of research on the integration of machine learning and optimization. One expansive area within this research stream is optimization over pre-trained predictive models, which proposes the use of pre-trained predictive models as surrogates for uncertain or highly complex objective functions. In this setting, features of the predictive models become decision variables in the optimization problem. Despite a recent surge in publications in this area, only a few papers note the importance of incorporating trust-region considerations in this decision-making pipeline, that is, enforcing solutions to be similar to the data used to train the predictive models. Without such constraints, the evaluation of the predictive model at solutions obtained from optimization cannot be trusted and the practicality of the solutions may be unreasonable. In this paper, we provide an overview of the approaches appearing in the literature to construct a trust region and propose three alternative approaches. Our numerical evaluation highlights that trust-region constraints learned through our newly proposed approaches compare favorably with previously suggested approaches, both in terms of solution quality and computational time. History: Accepted by Pascal Van Hentenryck, Area Editor for Computational Modeling: Methods & Analysis. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0312 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0312 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .