Wouter Jongeneel, Daniel Kuhn 0001, Mengmeng Li. A large deviations perspective on policy gradient algorithms. In Alessandro Abate, Mark Cannon, Kostas Margellos, Antonis Papachristodoulou, editors, 6th Annual Learning for Dynamics & Control Conference, 15-17 July 2024, University of Oxford, Oxford, UK. Volume 242 of Proceedings of Machine Learning Research, pages 916-928, PMLR, 2024. [doi]
Abstract is missing.