Chapter
AlphaGo Zero and the Limits of Traditional AI
1:05:19 - 1:10:28 (05:09)
David Silver, a leading researcher in AI and one of the creators of AlphaGo, discusses how traditional AI approaches hit their limit with complex games like Go and how AlphaGo Zero and AlphaZero surpassed these limitations through self-play and reinforcement learning.
Clips
Self play as a concept was first introduced in backgammon research by Gerald Tesauro's paper on training an expert using self play.
1:05:19 - 1:05:58 (00:38)
Artificial Intelligence
Summary
Self play as a concept was first introduced in backgammon research by Gerald Tesauro's paper on training an expert using self play. This term later gained traction in AI research and has become a more general and powerful concept.
ChapterAlphaGo Zero and the Limits of Traditional AI
Episode#144 – Michael Littman: Reinforcement Learning and the Future of AI
PodcastLex Fridman Podcast
Experts discuss the limitations and breakthroughs of neural nets in gaming, highlighting the success of AlphaGo and AlphaZero, which beat world-class players without any supervised learning on expert games, using only self-play.
1:05:58 - 1:10:28 (04:30)
Neural Networks
Summary
Experts discuss the limitations and breakthroughs of neural nets in gaming, highlighting the success of AlphaGo and AlphaZero, which beat world-class players without any supervised learning on expert games, using only self-play.