Poker Coverage: Poker Legislation Poker Tournaments Daily Fantasy Sports 2016 WSOP
Wsopbanner

Computer Beats World's Top Go Player In Historic Match

Google DeepMind’s AlphaGo Does Drop One Game To Lee Sedol

Print-icon
 

usgo.orgComputers cannot yet beat the world’s top poker pros, but they can dominate the world’s most elite Go players.

This week, a historic match in South Korea between Google DeepMind’s AlphaGo and world champion Lee Sedol ended with the machine winning four games to the human’s one.

Lee has been described as the “Roger Federer of Go,” according to theverge.com.

Go is a 2,500-year-old Chinese board game that many regard as the most complex board game in the world. Go played on the standard 19-by-19 grid has a state space of 10^171, while the game of chess has 10^50. Both games haven’t been solved, even though machines beat top humans.

In Go, players have roughly 200 possible moves to chose from, roughly ten times the number typically available in chess. Unlike chess, it’s challenging at times to determine who is winning in Go until the game reaches it later stages.

“Go has always been the pinnacle of perfect information games,” Demis Hassabis, co-founder of the company that developed AlphaGo, told The Verge. “It’s way more complicated than chess in terms of possibility, so it’s always been a bit of a holy grail or grand challenge for AI research, especially since Deep Blue. And you know, we hadn’t got that far with it, even though there’d been a lot of efforts. Monte Carlo tree search was a big innovation ten years ago, but I think what we’ve done with AlphaGo is introduce with the neural networks this aspect of intuition, if you want to call it that, and that’s really the thing that separates out top Go players: their intuition.”

Google’s AlphaGo crushed the European champ in January by a 5-0 score.

That Lee was able to get the machine to resign in a game is regarded as a huge achievement.

The AlphaGo system, developed by British computer company DeepMind, was bought by Google in 2014, according to the BBC. DeepMind was founded just six years ago.

What’s next for games-playing AI? Poker could be the answer.

“No-limit poker is very difficult,” Hassabis said.