TD-Gammon
Software / App
A computer backgammon program that learned to play at a world-class level using temporal difference reinforcement learning and self-play.
Mentioned in 2 videos
Save the 2 videos on TD-Gammon to your own pod.
Sign up free to keep building your knowledge base on TD-Gammon as more episodes are added.
Videos Mentioning TD-Gammon
Michael Littman: Reinforcement Learning and the Future of AI | Lex Fridman Podcast #144
Lex Fridman
A computer backgammon program that learned to play at a world-class level using temporal difference reinforcement learning and self-play.
Ilya Sutskever: OpenAI Meta-Learning and Self-Play | MIT Artificial General Intelligence (AGI)
Lex Fridman
An early application of self-play using Q-learning and neural networks to play backgammon, developed in 1992.