Offline reinforcement learning (RL) can learn control policies from static datasets but, like
standard RL methods, it requires reward annotations for every transition. In many cases …

Offline reinforcement learning (RL) algorithms have shown promising results in domains
where abundant pre-collected data is available. However, prior methods focus on solving …

Progress in deep learning highlights the tremendous potential of utilizing diverse robotic
datasets for attaining effective generalization and makes it enticing to consider leveraging …

Consider the following instance of the Offline Meta Reinforcement Learning (OMRL)
problem: given the complete training logs of $ N $ conventional RL agents, trained on $ N …

Recently, diffusion model shines as a promising backbone for the sequence modeling
paradigm in offline reinforcement learning (RL). However, these works mostly lack the …

Reinforcement learning (RL) algorithms hold the promise of enabling autonomous skill
acquisition for robotic systems. However, in practice, real-world robotic RL typically requires …

Cooperative multi-agent reinforcement learning (MARL) faces the challenge of adapting to
multiple tasks with varying agents and targets. Previous multi-task MARL approaches …

Reinforcement learning is a promising paradigm for solving sequential decision-making
problems, but low data efficiency and weak generalization across tasks are bottlenecks in …

Generalization and sample efficiency have been long-standing issues concerning
reinforcement learning, and thus the field of Offline Meta-Reinforcement Learning (OMRL) …

Offline meta-reinforcement learning (OMRL) utilizes pre-collected offline datasets to
enhance the agent's generalization ability on unseen tasks. However, the context shift …