Research

My research area lies at the intersection of Distributionally Robust Optimization and Mathematical Risk Theory. In particular, I study stochastic decision problems in a risk-averse environment, where decisions are evaluated using a risk measure . These types of problems are frequently encountered in finance (e.g., portfolio optimization), operations research (e.g., inventory management with uncertain demand), and machine learning (e.g., robust regression).

Over the past few years, my research has been focused on developing efficient optimization and statistical techniques to obtain and analyze solutions of (robust) risk-averse stochastic optimization problems. The outputs are listed below. A research statement can be found here.

Preprints

A New Distributionally Robust Optimization Model Based on Maxiance Regularization

Joint work with Roger J. A. Laeven and Henry Lam

GitHub

Constructing Uncertainty Sets for Robust Risk Measures: A Composition of $\phi$-Divergences Approach to Combat Tail Uncertainty

Joint work with Roger J. A. Laeven, Dick den Hertog and Aharon Ben-Tal

GitHub

Robust Optimization of Rank-Dependent Models with Uncertain Probabilities

Joint work with Roger J. A. Laeven and Dick den Hertog

GitHub

Submitted to Management Science

ROBIST: Robust Optimization by Iterative Scenario Sampling and Statistical Testing

Joint work with Justin Starreveld, Roger J. A. Laeven and Dick den Hertog

GitHub

Under Revision at Computers & Operations Research

Work in Progress

Sample Average Approximation of Risk Functionals: Non-Asymptotic Error Bounds

Joint work with Roger J. A. Laeven and Volker Kraetschmer