Distributionally Robust Learning and Optimization
Robust statistical procedures often lead to minimax, nonconvex, or large-scale optimization problems that are difficult to solve with generic algorithms. My work studies the structure created by distributional robustness, invariance, and heterogeneous data, and uses it to design computational methods that make these statistical targets practical without losing their interpretation or theoretical guarantees.
Related work 3
-
Koo, T. and Guo, Z. (2025).
Distributionally Robust Synthetic Control: Ensuring Robustness Against Highly Correlated Controls and Weight Shifts.
Technical Report
-
Guo, Z., Wang, Z., Hu, Y., and Bach, F. (2025).
Statistical Analysis of Conditional Group Distributionally Robust Optimization with Cross-Entropy Loss.
Technical Report
-
Wang, Z.#, Hu, Y.#, Bühlmann, P.✉, and Guo, Z.✉ (2024).
Causal Invariance Learning via Efficient Nonconvex Optimization.
Technical Report