Chapter
AlphaZero learns to play games: Men still needed.
1:05:58 - 1:21:46 (15:48)
The main idea of AlphaZero is to come up with a single elegant principle by which a system can learn for itself all of the knowledge which it requires to play a game such as Go, without any human intervention. Learning through reinforcement and deep learning, the things one figures out are actually be applicable to other problems that are real world problems.
Clips
The interviewee discusses the impact of AI technology in computer chess and Go, noting the victories of Deep Blue, AlphaGo, and AlphaZero over world champions, and including reflections on the inspiring stories that arise from such defeats.
1:05:58 - 1:10:49 (04:51)
Summary
The interviewee discusses the impact of AI technology in computer chess and Go, noting the victories of Deep Blue, AlphaGo, and AlphaZero over world champions, and including reflections on the inspiring stories that arise from such defeats.
ChapterAlphaZero learns to play games: Men still needed.
Episode#86 – David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning
PodcastLex Fridman Podcast
The limitations in machine learning occur when knowledge is put into a system, which then becomes difficult to extricate from that system.
1:10:49 - 1:18:46 (07:57)
Summary
The limitations in machine learning occur when knowledge is put into a system, which then becomes difficult to extricate from that system. This is where AlphaZero's aim to create a single elegant principle by which a system can learn for itself all the knowledge it requires separates itself by stepping back from knowledge integration.
ChapterAlphaZero learns to play games: Men still needed.
Episode#86 – David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning
PodcastLex Fridman Podcast
Former world champion of Go, Demis Hassabis, discusses the scientific advancements made in the development of AlphaGo and its ability to surpass previous versions of the AI technology.
1:18:46 - 1:21:46 (02:59)
Summary
Former world champion of Go, Demis Hassabis, discusses the scientific advancements made in the development of AlphaGo and its ability to surpass previous versions of the AI technology.