5 papers
Optimal Transport Meets Reinforcement Learning: A Survey
Yujie Zhu, Charles A. Hepburn, Matthew Thorpe +1
Reinforcement learning (RL) algorithms frequently compare probability distributions, such as state visitation distributions induced by policies and experts, action distributions fr…
Uncertainty-Based Smooth Policy Regularisation for Reinforcement Learning with Few Demonstrations
Yujie Zhu, Charles A. Hepburn, Matthew Thorpe +1
In reinforcement learning with sparse rewards, demonstrations can accelerate learning, but determining when to imitate them remains challenging. We propose Smooth Policy Regularisa…
State-Constrained Offline Reinforcement Learning
Charles A. Hepburn, Yue Jin, Giovanni Montana
Traditional offline reinforcement learning (RL) methods predominantly operate in a batch-constrained setting. This confines the algorithms to a specific state-action distribution p…
Model-based trajectory stitching for improved behavioural cloning and its applications
Charles A. Hepburn, Giovanni Montana
Behavioural cloning (BC) is a commonly used imitation learning method to infer a sequential decision-making policy from expert demonstrations. However, when the quality of the data…
Model-based Trajectory Stitching for Improved Offline Reinforcement Learning
Charles A. Hepburn, Giovanni Montana
In many real-world applications, collecting large and high-quality datasets may be too costly or impractical. Offline reinforcement learning (RL) aims to infer an optimal decision-…