method · first seen 19 Oct 2026
posterior sampling reinforcement learning
Also: PSRL, Thompson sampling for RL
Reinforcement learning that samples an MDP from a posterior and acts optimally for the sample.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Towards instance-dependent regret optimality in Episodic MDPs with Posterior SamplingPapers · 19 Oct 2026 · Victor Boone, Cyrille Kone, Waris Radji, Odalric-Ambrym Maillard, Dorian Baudry
Latest sources
- Towards instance-dependent regret optimality in Episodic MDPs with Posterior SamplingPapers · 19 Oct 2026 · Victor Boone, Cyrille Kone, Waris Radji, Odalric-Ambrym Maillard, Dorian Baudry