← Search

Soumith Udatha

2 accepted papers

2023

Reinforcement Learning with Probabilistically Safe Control Barrier Functions for Ramp Merging

ICRA 2023poster

Prior work has looked at applying reinforcement learning (RL) approaches to autonomous driving scenarios, but the safety of the algorithm is often compromised due to instability or the presence of ill-defined reward functions. With the use of control barrier functions embedded into the RL policy, we…

Cited by 11SourceScholar
2022

Imitating Past Successes can be Very Suboptimal

NeurIPS 2022accept

Prior work has proposed a simple strategy for reinforcement learning (RL): label experience with the outcomes achieved in that experience, and then imitate the relabeled experience. These outcome-conditioned imitation learning methods are appealing because of their simplicity, strong performance, an…

Cited by 19SourcePDFScholar