Go

AlphaGo defeats Lee Sedol (2016): learned intuition in Go

DeepMind combined policy/value nets with Monte Carlo Tree Search to beat a Go champion in a huge state space.