Philip Amortila

Philip Amortila is a postdoctoral scholar in the Machine Learning Research Pod at the Simons Institute for the Theory of Computing at UC Berkeley. He is delighted to be hosted by Peter Bartlett, Song Mei, and Jason Lee.

He completed his PhD at the University of Illinois Urbana-Champaign under the supervision of Nan Jiang. His research has focused on achieving sound and useful algorithms for reinforcement learning (RL) in modern settings, including those necessitating function approximation and representation learning. He has interned with the RL groups at Microsoft Research, Amazon Research, and the University of Alberta. He has been supported by an NSERC Doctoral Fellowship, and his work has been recognized a student best paper award at ALT 2021, an oral presentation at NeurIPS 2024, and a spotlight presentation at ICLR 2024.