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

Dealing with sparse rewards is a long-standing challenge in reinforcement learning (RL).
Hindsight Experience Replay (HER) addresses this problem by reusing failed trajectories for …

In goal-conditioned reinforcement learning (GCRL), sparse rewards present significant
challenges, often obstructing efficient learning. Although multi-step GCRL can boost this …

Offline reinforcement learning (RL) bears the promise to learn effective control policies from
static datasets but is thus far unable to learn from large databases of heterogeneous …

It is a challenge to make robots move from one place to another on the shortest path and
avoid obstacles at the same time, especially when there are some special conditions occur …

Reinforcement learning (RL) provides a formalism for learning-based control. By attempting
to learn behavioral policies that can optimize a user-specified reward function, RL methods …

Recent advances in machine learning using deep neural networks have shown significant
successes in learning from large datasets. However, these successes concentrated on …

Many recent offline RL algorithms attain both good empirical performance and enjoy
theoretical guarantees, but their applicability is limited to settings where data is collected for …