• Adversarial Shapley Value Experience Replay for Task-Free Continual Learning (University of Toronto)
• An Empirical Study of Human Behavioral Agents in Bandits, Contextual Bandits and Reinforcement Learning (IBM, Columbia University, Mila, Universite de Montreal)
• Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess (DeepMind, Vladimir Kramnik)
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets (UC Berkeley, this is actually from June but the write-up here was released this month)
• COVID-19 Pandemic Cyclic Lockdown Optimization Using Reinforcement Learning (Oracle)
• Decoupling Representation Learning from Reinforcement Learning (with Pieter Abbeel)
• GLIB: Efficient Exploration for Relational Model-Based Reinforcement Learning via Goal-Literal Babbling (MIT)
• Grounded Language Learning Fast and Slow (DeepMind, follow-up to SHIFTT)
• Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning (Russ Tedrake)
• Learning to summarize from human feedback (OpenAI)
• Meta-Learning with Sparse Experience Replay for Lifelong Language Learning (Facebook, King's College London, University of Amsterdam)
• Physically Embedded Planning Problems: New Challenges for Reinforcement Learning (DeepMind)