Data-driven optimization: predict, then optimize – Adil Bagirov (Federation University Australia, Ballarat, Victoria, Australia)

In many real-world applications of operations research machine learning and optimization are combined to make decisions. Typically, machine learning tools are used to predict or estimate key unknown parameters of the optimization models and then the optimization models are used to generate decisions. Such an approach leads to the development of the predict-then-optimize paradigm. This approach involves stochastic programming and nonsmooth optimization models and methods. In this talk we will discuss some of those models and methods.