Provably Efficient Algorithms for Large-Scale Reinforcement Learning

Livrables

The Open Research Data Pilot will be prepared and submitted to the European Commission

Publications

Efficient Global Planning in Large MDPs via Stochastic Primal-Dual Optimization

Auteurs: Gergely Neu, Nneka Okolo
Publié dans: Proceedings of The 34th International Conference on Algorithmic Learning Theory (ALT 2023), 2023
Éditeur: Proceedings of Machine Learning Research

Lifting the Information Ratio: An Information-Theoretic Analysis of Thompson Sampling for Contextual Bandits

Auteurs: Gergely Neu, Julia Olkhovskaya, Matteo Papini, Ludovic Schwartz
Publié dans: Advances in Neural Information Processing Systems 35 (NeurIPS 2022), 2022
Éditeur: NeurIPS foundation

Proximal Point Imitation Learning

Auteurs: Luca Viano, Angeliki Kamoutsi, Gergely Neu, Igor Krawczuk, Volkan Cevher
Publié dans: Advances in Neural Information Processing Systems 35 (NeurIPS 2022), 2022
Éditeur: NeurIPS foundation

Online learning with off-policy feedback

Auteurs: Germano Gabbianelli, Matteo Papini, Gergely Neu
Publié dans: Proceedings of The 34th International Conference on Algorithmic Learning Theory (ALT 2023), 2023
Éditeur: Proceedings of Machine Learning Research

Optimistic Planning by Regularized Dynamic Programming

Auteurs: Antoine Moulin, Gergely Neu
Publié dans: International Conference on Machine Learning (ICML 2022), 2023
Éditeur: Proceedings of Machine Learning Research

Generalization bounds via convex analysis

Auteurs: Gabor Lugosi, Gergely Neu
Publié dans: Proceedings of Thirty Fifth Conference on Learning Theory (COLT 2022), 2022
Éditeur: Proceedings of Machine Learning Research

Smoothing policies and safe policy gradients

Auteurs: Matteo Papini; Matteo Pirotta; Marcello Restelli
Publié dans: Machine Learning, Numéro 111, 2022, Page(s) 4081–4137, ISSN 1573-0565
Éditeur: Springer
DOI: 10.1007/s10994-022-06232-6

Livrables

Publications

Partager cette page

Télécharger