Kevin Li

Beyond Conservatism in Offline Reinforcement Learning: The Importance of Effective Representations

Kevin Li, 2022 .

Beyond Bayes-optimality: meta-learning what you know you don't know

Pedro A. Ortega, Jane X. Wang, J. Veness, 2022, ArXiv.

RobArch: Designing Robust Architectures against Adversarial Attacks

Duen Horng Chau, Cory Cornelius, Weilin Xu, 2023, ArXiv.

MURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning pdf

Sergey Levine, Aurick Zhou, Ashwin Reddy, 2021, ICML.

Robust Principles: Architectural Design Principles for Adversarially Robust CNNs

Duen Horng Chau, Cory Cornelius, Weilin Xu, 2023, British Machine Vision Conference.