Google DeepMind’s AlphaGo defeated Lee Sedol in a five-game Go match in 2016.
The match was held in Seoul, South Korea, from 9 to 15 March 2016. AlphaGo won four games to one, becoming the first computer Go program to defeat a top professional player in an even match under standard rules. Lee Sedol won the fourth game, which became famous for his creative play.
AlphaGo combined neural networks with tree search. Its policy network helped select promising moves, while its value network estimated positions. The system had previously trained on human games and then improved through games played against versions of itself.
Go has a much larger branching factor than chess, so older brute-force approaches struggled with it. AlphaGo’s victory was consequently seen as a major demonstration of combining learned pattern recognition with search, not as proof that the system understood the game like a human.