Dr Jekyll & Mr Hyde: the strange case of off-policy policy updates

von Laroche, Romain
Veröffentlicht in NeurIPS (35. : 2021 : Online) 35th Conference on Neural Information Processing Systems (NeurIPS 2021) ; Volume 29 of 36 (2022)

Safe Policy Improvement with Soft Baseline Bootstrapping

von Nadjahi, Kimia
Veröffentlicht in ECML PKDD (4. : 2019 : Würzburg) Machine learning and knowledge discovery in databases ; Part 3 (2020)

Multi-Objective SPIBB: Seldonian Offline Policy Improvement with Safety Constraints in Finite MDPs

von satija, harsh
Veröffentlicht in NeurIPS (35. : 2021 : Online) 35th Conference on Neural Information Processing Systems (NeurIPS 2021) ; Volume 3 of 36 (2022)

SAFE POLICY IMPROVEMENT WITH AN ESTIMATED BASELINE POLICY

von Simáo, Thiago D.
Veröffentlicht in AAMAS (19. : 2020 : Online) International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020) ; Volume 2 of 3 (2021)

When does return-conditioned supervised learning work for offline reinforcement learning?

von Brandfonbrener, David
Veröffentlicht in NeurIPS (36. : 2022 : New Orleans, La.; Online) 36th Conference on Neural Information Processing Systems (NeurIPS 2022 ; Volume 3 of 50 (2023)

Discrete Compositional Representations as an Abstraction for Goal Conditioned Reinforcement Learning

von Islam, Riashat
Veröffentlicht in NeurIPS (36. : 2022 : New Orleans, La.; Online) 36th Conference on Neural Information Processing Systems (NeurIPS 2022 ; Volume 6 of 50 (2023)

LEARNING DYNAMIC BELIEF GRAPHS TO GENERALIZE ON TEXT-BASED GAMES

von Adhikari, Ashutosh
Veröffentlicht in NeurIPS (34. : 2020 : Online) 34th Conference on Neural Information Processing Systems (NeurIPS 2020) ; Volume 4 of 27 (2021)