Tea Time Talks 2024: Yiuqi Wang, Transformers Learn Temporal Difference Methods for In-Context RL
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:
Abstract: In this talk, I will discuss how transformers Can and Do implement in-context temporal difference (TD) learning. In-context learning refers to the learning ability of a model during inference time without adapting its parameters --- a phenomenon observed in deep learning models, especially in large language models based on the transformer architecture. We show that transformers implement TD in the context under some special weights. In addition, training the transformer with a multi-task TD algorithm converges with the weight that implements in-context TD.
Смотрите видео Tea Time Talks 2024: Yiuqi Wang, Transformers Learn Temporal Difference Methods for In-Context RL 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 40 раз и оно понравилось 3 людям.