Q Daily | Human score 1 vs. Alpha Go score 4 — lost to artificial intelligence

March 20, 2016

Q Daily — March 15, 2016 | Maruo Fei Cuiqi Wen

This is a summary. Read original article in full here.

After a week of the man vs. machine tournament of go, the reigning world go champion Lee Sedol faced off for the final match against the artificial intelligence system called Alpha Go. Lee Sedol only scored a total of 1:4 — losing to Google’s AI algorithm Alpha Go.

Another artificial intelligence called Deep Blue, built by IBM, was used 20 years ago to beat the world’s chess champion, spelling the beginning of machine learning’s rivalry with human ability. What does this new artificial intelligence victory mean for the future of AI?

James Barrett explores this question in his book Our Final Invention: Artificial Intelligence and the End of the Human Era in which he interviews futurist Ray Kurzweil.

Alpha Go is artificial intelligence in a narrow sense, it can only be used to play go. But our brains can do many different tasks, humans have a universal intelligence. However, this development can lead to the creation of a robot, a computer system that has farther reaching abilities, this depth of learning is a very important step.


related reading:
Google | Deep Mind: main
Google | Deep Mind: Alpha Go
Google | Alpha Go using machine learning to master the ancient game of go: summary

Nature | main
Nature | Mastering the game of go with deep neural networks and tree search: paper

Deep Mind | Alpha Go
About the Nature paper.

Our Nature paper describes the technical details behind a new approach to computer go that combines Monte Carlo tree search with deep neural networks that have been trained by supervised learning, from human expert games, and by reinforcement learning from games of self-play.

The game of go is widely viewed as an unsolved “grand challenge” for artificial intelligence. Despite decades of work, the strongest computer go programs still only play at the level of human amateurs. In this paper we describe our go program, Alpha Go. This program was based on general purpose AI methods, using deep neural networks to mimic expert players, and further improving the program by learning from games played against itself.

Alpha Go won over 99% of games against the strongest other go programs. It also defeated the human European champion by 5–0 in an official tournament match. This is the first time ever that a computer program has defeated a professional go player, a feat previously believed to be at least a decade away.


video | How AI beat the world’s top human at Chinese board game of go
Background on the historic artificial intelligence victory.

Nature | Go is an ancient Chinese board game, often viewed as the game computers could never play. Now researchers from Google owned company Deep Mind have proven the nay-sayers wrong, creating an artificial intelligence, called Alpha Go, which has beaten a professional go player for the first time.

In this Nature video, we go behind the scenes to learn about the game, the program and what this means for the future of artificial intelligence.

related reading:
Nature | YouTube channel: main
Nature | YouTube channel: news stream