401S Bridge Hall
I am an Assistant Professor at USC Marshall Data Sciences and Operations, in the Operations group.
Previously I was a research fellow at the Simons program on causality, a FODSI postdoc at UC Berkeley, hosted by Bin Yu, and Michael I. Jordan. I obtained my PhD from Cornell University in Operations Research and Information Engineering working with Nathan Kallus at Cornell Tech. My work was previously supported on a NDSEG fellowship.
My research interests are broadly in data-driven decision making under uncertainty, including operations, statistical machine learning, and causal inference, and the interplay of statistics and optimization.
Email: zhoua at usc.edu.
Research interests by topic:
- Computerized influence functions for optimization-based estimators in causal inference
- Program evaluation perspective on algorithmic accountability; equity and efficacy in the provision of social services: fair/optimal encouragement designs
- Credible causal inference for optimal decisions and bounds; and for infinite-horizon and finite horizon (scalable) offline reinforcement learning under violations of assumptions
- Algorithmic fairness: biased data, partial identification bounds on causal disparities. Substantive: don’t use CJ data for ML benchmarks, evaluating short term impacts of bail reform on aggregate crime.
- Data-Driven Influence Functions for Causal Inference and Optimization-Based EstimatorsSupersedes Empirical Gateaux Derivatives for Causal Inference (oral at Neurips 2022)
- Robust Fitted-Q-Evaluation and Iteration under Sequentially Exogenous Unobserved Confounders2023
- Optimizing and Learning Sequential Assortment Decisions with Platform Disengagement2023
- Minimax-Optimal Policy Learning under Unobserved ConfoundingManagement Science (2021), supersedes Neurips 2018 version 2023
- Assessing algorithmic fairness with unobserved protected class using data combinationManagement Science (2020). A preliminary version appeared at FaCCT 2020 2023
|Sep 21, 2023||Delighted to share that my paper on Optimal and Fair Encouragement Policy Evaluation and Learning is accepted at Neurips 2023! TL;DR: under incomplete take-up, we care both about utility and who gets access to services. Let’s work together to equitably reduce administrative burdens. See you in New Orleans!|
|Sep 21, 2023||Big update to Robust Fitted-Q-Evaluation and Iteration under Sequentially Exogenous Unobserved Confounders, with new and exciting results on complex healthcare data and warm-starting RL!|
|Aug 30, 2023||Check out our report on EAAMO 2022 in SIGecom Exchanges!|
|Aug 23, 2023||I’ll be an Area Chair for AISTATS 2023.|
|Aug 10, 2023||New paper posted on Optimizing and Learning Assortment Decisions in the Presence of Platform Disengagement! (with Mika Sumida)|
|Jul 1, 2023||Our paper on An Empirical Evaluation of the Impact of New York’s Bail Reform on Crime Using Synthetic Controls was accepted at the journal of Statistics and Public Policy!|
|Jun 1, 2023||Major (working) update on Data-Driven Influence Functions for Causal Inference and Optimization-Based Estimators, with new results on sensitivity analysis!|
|Apr 13, 2023||Upcoming talks at USC ShowCAIS (April), USC Econometrics Reading group (April), Simons (Causality reunion) (May), Socal OR/OM Day (May), Stanford OIT (June), MSOM|
|Feb 1, 2023||New paper on Robust Fitted-Q-Iteration under Unobserved Confounders!|
|Jan 15, 2023||Serving as a Tutorials Chair and an Area-Chair for FAccT 2023, Scientific Integrity Chair for UAI 2023.|