In Machines We Trust: How games teach AI to learn for itself (Update)
MIT Technology Review Narrated
00:00
AlphaGo - The First AI to Play Board Games at the Level of Top Professionals
AlphaGo was the first program to be able to play Go at the level of top human professional players. Silver's next system, AlphaZero, learned to play board games and video games in a different way by learning the rules of a game and then playing itself over and over again. It is a milestone in making AI more general purpose, but with an important caveat. The algorithm can't learn to play these games all at once - it has to swap out its chest brain before it can play Go.
Transcript
Play full episode
