Rate My Professor Maryam Fazel

Her research focuses on optimization in machine learning and AI, deep learning theory, learning and control, and reinforcement learning. Fazel has secured major grants as principal investigator or co-principal investigator, including DARPA's Lagrange program for control and learning of uncertain dynamical systems (2018), NSF CIF grants for mathematical foundations of deep reinforcement learning (2022–2026) and learning with budget constraints (2020–2024), and NSF AF grant for machine learning markets (2023–2026). Notable awards include the 2025 Farkas Prize from the INFORMS Optimization Society, NSF CAREER Award (2009), University of Washington Electrical Engineering Outstanding Teaching Award (2009), and UAI Best Student Paper Award (2014). Key publications encompass 'Extragradient Preference Optimization (EGPO): Beyond Last-Iterate Convergence for Nash Learning from Human Feedback' (arXiv 2025), 'Flat Minima Generalize for Low-Rank Matrix Recovery' (IMA Journal of Information and Inference, 2024), 'Global Convergence of Policy Gradient for Linear Quadratic Regulator' (2018), and 'Hankel Matrix Rank Minimization with Applications to System Identification and Realization' (2011), which was recognized as a Fast Breaking Paper by ScienceWatch. She contributes editorially as Founding Associate Editor of SIAM Journal on Mathematics of Data Science, Program Chair of ICML 2025, Action Editor for Journal of Machine Learning Research, and on the MOS-SIAM Optimization Book Series board. Her influence is evident in mentoring alumni to positions at Google DeepMind, universities, and industry, and organizing cross-campus seminars in optimization and data science.