Postdoctoral Researcher
Department of Electrical Engineering and Computer Sciences
UC Berkeley
I am a postdoctoral researcher at UC Berkeley, hosted by Michael I. Jordan. Previously, I was a PhD student at ETH Zurich in Florian Dörfler's research group. My current research lies at the intersection of optimization, statistics, and computation, focusing on the development of principled and computationally tractable methods for decision-making under uncertainty. I am particularly interested in complex forms of uncertainty, such as distribution shifts, rare events (e.g., black swans), and structured uncertainty (e.g., over networks). My theoretical research is generally motivated by fundamental problems in machine learning, energy markets, and networked systems, where uncertainty is central to reliable decision-making and system resilience. My PhD research focused primarily on developing the foundations of optimal transport-based distributionally robust optimization. Earlier in my PhD, I studied the stability and robustness of nonlinear dynamical systems, with applications to renewable-integrated power systems.
(1) Stochastic Optimization; (2) Robust Optimization; (3) Anytime-Valid Sequential Inference; (4) Uncertainty Quantification for AI
[1] “Stopping Rules for Stochastic Gradient Descent via Anytime-Valid Confidence Sequences”. Under Review in Mathematics of Operations Research. ArXiv version: arxiv:2512.13123. With Michael I. Jordan.
[2] “Diffusion-Robust Optimization over Graphs”. Under Review in Mathematics of Operations Research. ArXiv version: arxiv:2605.30853. With Ricky Huang, Paul Grigas, Michael I. Jordan.
[3] “Stochastic Optimization with Optimal Importance Sampling”. Under Review in Mathematics of Operations Research. ArXiv version: arxiv:2504.03560. With B. Van Parys, H. Lam, M.I. Jordan.
[1] “Wasserstein Distributionally Robust Estimation in High Dimensions: Performance Analysis and Optimal Hyperparameter Tuning”. Accepted in Mathematical Programming (Series A). ArXiv version: arxiv:2206.13269. With S. Shafiee, F. Dörfler.
[2] “Nash Equilibria, Regularization and Computation in Optimal Transport-Based Distributionally Robust Optimization”. Accepted in Operations Research. ArXiv version: arXiv:2303.03900. With S. Shafiee, F. Dörfler, D. Kuhn.
[3] “Revisiting mean estimation over ℓp balls: Is the MLE optimal?”. Major Revision in The Annals of Statistics. ArXiv version: arxiv:2506.10354. With M.I. Jordan, R. Pathak, A. Ulichney.
[4] “Minimum Volume Conformal Sets for Multivariate Regression”. Accepted in Journal of the American Statistical Association (Theory and Methods). ArXiv version: arxiv:2503.19068. With S. Braun, M.I. Jordan, F. Bach.
[5] “Conformal Prediction under Levy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations". Accepted in Advances in Neural Information Processing Systems. ArXiv version: arxiv:2502.14105. With Z.O. Wang, J. Zhu, M.I. Jordan, Y. Marzouk.
[6] “Distributional Uncertainty Propagation via Optimal Transport”. Accepted in IEEE Transactions on Automatic Control. ArXiv version: arxiv:2205.00343. With N. Lanzetti, H. Chen, F. Dörfler.
[1] “Hedging Against Black Swans in Day-Ahead Energy Markets". Accepted in Power Systems Computation Conference (PSCC) + Electric Power Systems Research (EPSR). ArXiv version: arxiv:2510.14328. With B. Bangoura, S. Bolognani, N. Lanzetti, F. Dörfler.
Contact: liviu.aolaritei@berkeley.edu