method · first seen 29 Oct 2026
minimax-optimal regret bound
Regret guarantees in reinforcement learning that match the information-theoretic lower bound up to constants.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Closing the Log Gap: Minimax-Optimal Regret for Episodic RLPapers · 29 Oct 2026 · Guy Zamir, Victor Boone, Matthew Zurek, Yudong Chen
Latest sources
- Closing the Log Gap: Minimax-Optimal Regret for Episodic RLPapers · 29 Oct 2026 · Guy Zamir, Victor Boone, Matthew Zurek, Yudong Chen