Simon S. Du is an Assistant Professor at the Paul G. Allen School of Computer Science & Engineering at the University of Washington. Prior to starting as faculty, he was a postdoc at the Institute for Advanced Study of Princeton, under the mentorship of Sanjeev Arora. He completed his Ph.D. in Machine Learning at Carnegie Mellon University, where he was advised by Aarti Singh and Barnabás Póczos. Simon’s research has been recognized by a Sloan Research Fellowship, a Samsung AI Researcher of the Year Award, an Intel Rising Star Faculty Award, an NSF CAREER award, an Nvidia Pioneer Award, a Distinguished Dissertation Award honorable mention from CMU, among others. His notable contributions include proving the first global convergence result of gradient descent for optimizing deep neural networks, settling the sample complexity in reinforcement learning, and establishing the necessary and sufficient conditions for reinforcement learning in large state spaces. His current research focuses multi-agent reinforcement learning and data selection algorithms for foundation models.

AI2050 Project

This project aims to develop AI systems that can effectively cooperate with humans, a crucial challenge as AI becomes more integrated into our lives. Unlike existing AI systems, which struggle to adapt to new human partners, our research focuses on creating theoretical foundations and new algorithms that enable AI to work seamlessly with people in diverse situations. We aim to build AI systems that collaborate with us in meaningful, practical ways, paving the way for more beneficial AI technologies in the future.