Simon Shaolei Du

Simon Shaolei Du is an Assistant Professor at the Paul G. Allen School of Computer Science & Engineering, University of Washington. His research focuses on machine learning, spanning deep learning, representation learning, reinforcement learning, and data selection. He previously held postdoctoral positions at the Institute for Advanced Study of Princeton and has collaborated with major research labs including Meta, Google, and Microsoft.

• Ph.D. in Machine Learning, Carnegie Mellon University (2019)
• Co-advised by Aarti Singh and Barnabás Póczos
• Undergraduate study in EECS and EMS at UC Berkeley

Du’s research addresses foundational questions in machine learning, including optimization guarantees for neural networks, sample complexity in reinforcement learning, and theoretical insights into preference learning frameworks like RLHF and DPO. His work bridges practical algorithm design with rigorous mathematical analysis.

Recent publications highlight his expertise in reinforcement learning (e.g., policy-based trajectory clustering, latent state decoding), optimization theory (e.g., gradient descent dynamics, saddle point escape), and representation learning (e.g., overparameterized Gaussian mixtures). These span both theoretical and applied dimensions of machine learning.

• IEEE AI’s 10 to Watch, 2024
• Schmidt Sciences AI2050 Early Career Fellow, 2024
• Alfred P. Sloan Research Fellowship, 2024
• Intel Rising Star Faculty Award, 2023
• Samsung AI Researcher of the Year, 2022
• NSF CAREER Award, 2022

Du’s affiliations include the NSF AI Institute for Societal Decision Making (AI-SDM) and the Simons Institute for Theoretical Computer Science. He actively mentors researchers and welcomes prospective students and postdocs via email submissions with CVs.