pgradu@berkeley.edu
I’m a fifth year PhD candidate in EECS at UC Berkeley advised by Ben Recht and Michael Jordan. Before, I was very fortunate to be advised by Elad Hazan (at Princeton) who taught me the ways of research.
My work integrates techniques from online optimization, modern statistics and control theory and is focused on developing adaptive methods for evaluating interventions and individualizing interventions with provable guarantees. For the former, examples include: identifying disparities across subgroups in incident databases, proposing a variance-optimal sequential experiment design, and designing methodology for valid post-causal-discovery inference. For the latter, a key example is designing an individualized approach to tapering medications, and on-going work with UCSF on developing an optimization framework for dose reductions in chemotherapy.
From Individual Experience to Collective Evidence: An Incident-Based Framework for Identifying Systemic Discrimination ICML 2025. With Jessica Dai, Deborah Raji, and Ben Recht.
Online Control For Adaptive Tapering Of Medications CDC 2023. With Ben Recht.
CLIP-OGD: An Experimental Design for Adaptive Neyman Allocation in Sequential Experiments NeurIPS 2023. With Jessica Dai, and Chris Harshaw.
Valid Inference after Causal Discovery. JASA 2024. With Tijana Zrnic, Yixin Wang, and Michael I. Jordan.
Adaptive Regret for Control of Time-Varying Dynamics. L4DC 2023. With Elad Hazan, and Edgar Minasyan.
Projection-free Adaptive Regret with Membership Oracles ALT 2023. With Zhou Lu, Nataly Brukhim, and Elad Hazan.
Lyapunov Density Models: Constraining Distribution Shift in Learning-Based Control. ICML 2022. With Katie Kang, Jason Choi, Michael Janner, Claire Tomlin, and Sergey Levine.
Online Control of Unknown Time-Varying Dynamical Systems. NeurIPS 2021. With Edgar Minasyan, Max Simchowitz, and Elad Hazan.
Non-Stochastic Control with Bandit Feedback. NeurIPS 2020. With John Hallman, and Elad Hazan.
SyntaxNet for Romanian: Results and Potential. Proceedings of the 12th International Conference Linguistic Resources and Tools for Processing the Romanian Language, 2016. With Radu Ion.
Machine Learning for Mechanical Ventilation Control. (extended abstract) ML4H 2021. With Daniel Suo, Naman Agarwal, Wenhan Xia, Xinyi Chen, Udaya Ghai, Alexander Yu, Karan Singh, Cyril Zhang, Edgar Minasyan, Julienne LaChance, Tom Zajdel, Manuel Schottdorf, Daniel Cohen, and Elad Hazan.
DELUCA - A Differentiable Control Library: Environments, Methods, and Benchmarking. Differentiable Computer Vision, Graphics, and Physics in Machine Learning Workshop, NeurIPS 2020. With John Hallman, Daniel Suo, Alex Yu, Naman Agarwal, Udaya Ghai, Karan Singh, Cyril Zhang, Anirudha Majumdar, and Elad Hazan.