AlphaGo Documentary

[lpetrich]

AlphaGo - The Movie | Full Documentary - YouTube

Previous threads:
Artificial intelligence: more difficult than one might expect?
Some computer chess and Go champions -- they learned on their own
Computers become champions in yet another game - poker

Recently released by DeepMind:

With more board configurations than there are atoms in the universe, the ancient Chinese game of Go has long been considered a grand challenge for artificial intelligence. On March 9, 2016, the worlds of Go and artificial intelligence collided in South Korea for an extraordinary best-of-five-game competition, coined The DeepMind Challenge Match. Hundreds of millions of people around the world watched as a legendary Go master took on an unproven AI challenger for the first time in history.

Directed by Greg Kohs with an original score by Academy Award nominee, Hauschka, AlphaGo chronicles a journey from the halls of Oxford, through the backstreets of Bordeaux, past the coding terminals of DeepMind in London, and ultimately, to the seven-day tournament in Seoul. As the drama unfolds, more questions emerge: What can artificial intelligence reveal about a 3000-year-old game? What can it teach us about humanity?

 

[lpetrich]

I watched that documentary, and it showed how some software learned to play Breakout. First, moving the paddle so that the ball does not escape past it. Then, punching a hole and sending the ball to it so it can hit the bricks in the back.

Then on how difficult Go seems. Its rules are simple: each player places one stone at a time, and if any opponents' stones are surrounded, then they are removed. The winner is whoever has the most area. For a standard 19*19 board, the number of possible configurations is huge, and many players develop an intuition about how to play.

Then how it feels to be a very good Go player and lose to the software.

AlphaGo was first trained on existing games, by making it do what a good amateur might do, then it was trained against versions of itself.

We saw AG playing Go champion Lee Sedol, with AG winning in three games, and LS winning the fourth game, then AG winning the fifth game. LS had found a weak spot in AG's strategy, and he took advantage of it. A lot of people celebrated LS's 4th-game victory, as if he was standing up for our species.