Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observability
Dibya Ghosh, Jad Rahme, Aviral Kumar, Amy Zhang, Ryan P. Adams, Sergey Levine
People
Amy Zhang
PhD Student
I am a PhD candidate at McGill University in the Reasoning and Learning Lab supervised by Joelle Pineau and a Researcher at Meta Artificial Intelligence Research. I hold dual B.S. degrees in EECS and math, and an M.Eng. in EECS from MIT. I previously did research in computer vision, privacy-preserving recommendation, and network coding. I am also a board member of Women in Machine Learning.
Interests
I work on state abstractions and deep reinforcement learning, with a focus on model-based methods and generalization.
Latest Publications
arXiv - 2021
MBRL-Lib: A Modular Library for Model-based Reinforcement Learning
Luis Pineda, Brandon Amos, Amy Zhang, Nathan O. Lambert, Roberto Calandra
Multi-Task Reinforcement Learning with Context-based Representations
Shagun Sodhani, Amy Zhang, Joelle Pineau
Stable Policy Optimization via Off-Policy Divergence Regularization
Ahmed Touati, Amy Zhang, Joelle Pineau, Pascal Vincent
Learning Robust State Abstractions for Hidden-Parameter Block MDPs
Amy Zhang, Shagun Sodhani, Khimya Khetarpal, Joelle Pineau
Learning Invariant Representations for Reinforcement Learning without Reconstruction
Amy Zhang, Rowan McAllister, Roberto Calandra, Yarin Gal, Sergey Levine