We consider stochastic programs where the distribution of the uncertain parameters is only
observable through a finite training dataset. Using the Wasserstein metric, we construct a …

This is a substantial revision of the previous edition with added new material. The
presentation of Chapter 6 is updated. In particular the Interchangeability Principle for risk …

The goal of regression and classification methods in supervised learning is to minimize the
empirical risk, that is, the expectation of some loss function quantifying the prediction error …

In this tutorial we discuss several aspects of modeling and solving multistage stochastic
programming problems. In particular we discuss distributionally robust and risk averse …

Adaptive robust optimization problems are usually solved approximately by restricting the
adaptive decisions to simple parametric decision rules. However, the corresponding …

The planning of post-disruption restoration in critical infrastructure systems often relies on
deterministic assumptions such as complete information on resources and known duration …

In this paper, we study a data-enabled predictive control (DeePC) algorithm applied to
unknown stochastic linear time-invariant systems. The algorithm uses noise-corrupted …

We present a novel humanitarian supply chain approach to address disaster preparedness
and build response capacity in humanitarian supply chains when people's vulnerability …

Motivated by a leading taxi operator in Singapore, we consider the idle vehicle pre‐
allocation problem with uncertain demands and other uncertain covariate information such …

We present a general framework for robust satisficing that favors solutions for which a risk-
aware objective function would best attain an acceptable target even when the actual …