Reinforcement Learning
Learning Retrospective Knowledge with Reverse Reinforcement Learning Shangtong Zhang University of Oxford Vivek V eeriah University of Michigan, Ann Arbor Shimon Whiteson University of Oxford
We present a Reverse Reinforcement Learning (Reverse RL) approach for representing retrospective knowledge .
Widening the Pipeline in Human-Guided Reinforcement Learning with Explanation and Context-Aware Data Augmentation
Human explanation (e.g., in terms of feature importance) has been recently used
Joint Policy Search for Multi-agent Collaboration with Imperfect Information
To learn good joint policies for multi-agent collaboration with imperfect information remains a fundamental challenge.
Towards Minimax Optimal Reinforcement Learning in Factored Markov Decision Processes
We study minimax optimal reinforcement learning in episodic factored Markov decision processes (FMDPs), which are MDPs with conditionally independent transition components.
Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Models Minting Pan Xiangming Zhu Y unbo Wang
World models learn the consequences of actions in vision-based interactive systems. However, in practical scenarios such as autonomous driving, there commonly exists noncontrollable dynamics independent of the action signals, making it difficult to learn effective world models.
Accelerating Robotic Reinforcement Learning via Parameterized Action Primitives
These parameterized primitives are expressive, simple to implement, enable efficient exploration and can be transferred across robots, tasks and environments.
Reinforcement Learning with Augmented Data
Learning from visual observations is a fundamental yet challenging problem in Reinforcement Learning (RL).
Hierarchical Reinforcement Learning With Timed Subgoals
Hierarchical reinforcement learning (HRL) holds great potential for sample-efficient learning on challenging long-horizon tasks.