Reinforcement learning

The latest artificial intelligence systems start from zero knowledge of a game and grow to world-beating in a matter of hours. But researchers are struggling to apply these systems beyond the arcade.

A new version of AlphaGo needed no human instruction to figure out how to clobber the best Go player in the world — itself.

Computer scientists are finding ways to code curiosity into intelligent machines.

The Go-playing program teaches itself to replicate something very much like human intuition, an advance that promises far-reaching consequences.