My research formalizes model multiplicity to transform the existence of multiple predictive truths into a mechanism for safe, trustworthy, and interpretable AI. I develop the theoretical foundations to characterize Rashomon sets and the optimization algorithms required to navigate them for outcomes like provable recourse and personalized alignment. My work bridges the gap between machine learning theory and the practical needs of high-stakes decision-making in healthcare and policy.
Publications
Google Scholar | dblp | * denotes equal contribution-
2025
arXiv preprint arXiv:2511.21799, 2025arXiv preprint arXiv:2511.19636, 2025Advances in Neural Information Processing Systems (NeurIPS), 2025spotlightAdvances in Neural Information Processing Systems (NeurIPS), 2025Medical Image Computing and Computer Assisted Intervention (MICCAI), 2025An oral workshop version appeared at Medical Imaging meets NeurIPS WorkshopHarvard Data Science Review, 2025won 2022 American Statistical Association Data Challenge Expo Student Competition.A workshop version appeared at NeurIPS 2022 Workshop on Causality for Real-world ImpactJournal of the American Medical Informatics Association, 2025 -
2024
Advances in Neural Information Processing Systems (NeurIPS), 2024Proceedings of the International Conference on Machine Learning (ICML), 2024spotlightINFORMS Journal on Data Science, 2024Workshop on Interpretable Policies in Reinforcement Learning@ RLC-2024, 2024oral -
2023
Advances in Neural Information Processing Systems (NeurIPS), 2023The Journal of Infectious Disease (JID), 2023 -
2022
Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT), 2022Statistics Surveys, 2022 -
2021
Second Workshop on Scholarly Document Processing at NAACL, 2021oral, won third place in the 3C Shared Task Competition