Grainger College of Engineering, All Events

Abstract: With the emergence of big data, contextual information has become increasingly accessible to support data-driven decision-making under uncertainty, giving rise to a research area known as contextual stochastic optimization. In many real-world applications, the uncertainty could be impacted by both contextual information and the decisions to be optimized, leading to the so-called decision-dependent uncertainty. In this talk, we consider a residuals-based distributionally robust optimization model, where the underlying uncertainty depends on both contextual information and our decisions. We adopt both parametric and nonparametric regression models to learn the latent decision dependency and construct a nominal distribution (thereby ambiguity sets) around the learned model using empirical residuals from the regressions. We provide conditions under which desired statistical properties, such as asymptotic optimality, rates of convergence, and finite sample guarantees, are satisfied. To solve the resulting distributionally robust optimization model, we develop a customized Benders decomposition algorithm with nonlinear cuts and prove its finite convergence. Through numerical experiments, we illustrate the effectiveness of our approach and the benefits of integrating decision dependency into a residuals-based DRO framework.

Bio: Xian Yu is currently an Assistant Professor of Integrated Systems Engineering at The Ohio State University. She received her Ph.D. degree in Operations Research from the University of Michigan in 2022 and B.S. degree in Mathematics and Applied Mathematics from Xi’an Jiaotong University in 2017. Her research interest lies in sequential decision-making and optimization under uncertainty, with applications in transportation and logistics, supply chain management, and healthcare. Her work has been published in top-tier journals including Mathematical Programming, INFORMS Journal on Computing, and European Journal of Operational Research, as well as flagship AI and control conferences including ICML, AAAI, and IEEE CDC. Xian is the recipient of several awards, including the Honorable Mention in Dupačová-Prékopa Best Student Paper Prize in Stochastic Programming in 2025, the first place of IISE Pritsker Doctoral Dissertation Award in 2023, and Katta Murty Prize for the Best Student Research Paper on Optimization in 2021.