arXiv cs.AI / cs.LG / cs.CL·20d agoRetimed Bellman Flows: Escaping the Impossible Triangle of Velocity Bootstrapping#reinforcement-learning#flow-matching#offline-rlAI research
arXiv cs.AI / cs.LG / cs.CL·21d agoTrust Guided Decision Transformer#decision-transformer#reinforcement-learning#d4rlAI research