Tea Time Talks 2024: Alireza Kazemipour, Optimism and Mon-MDPs
Tea Time Talks are back for another year. This summer lecture series, presented by Amii and the RLAI Lab at the University of Alberta, give researchers the chance to discuss early-stage ideas and prospective research. Join us for another series of informal 20-minute talks where AI leaders discuss the future of machine learning research.
Abstract:
Arguably, the principle of Optimism in the Face of Uncertainty (OFU) was the primary driving force behind the development of many sample efficient exploration algorithms for reinforcement learning throughout the whole 2000s. However, works such as The End of Optimism in 2016 showed inefficiencies of optimism-based exploration methods obscured in their worst-case regret guarantees, and pointed toward more theoretically sound exploration principles such as Information-Directed Sampling as alternatives.
Nevertheless, in this talk I intend to present that under Mon-MDP formulation, OFU is not effective let alone efficient and I will try to spend time explaining what challenge OFU faces once applied to Mon-MDPs.
Смотрите видео Tea Time Talks 2024: Alireza Kazemipour, Optimism and Mon-MDPs онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Amii 27 Сентябрь 2024, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 35 раз и оно понравилось 0 людям.