HackerAlphaGo - The Movie | Full award-winning documentary
Google DeepMindYouTube ↗
Further reading 1
Caption notes
Follow AlphaGo from tests with Fan Hui to its match with Lee Sedol. The film shows how expert games, self-play and search meet the realities of tournament play.
Original passages · 5
AlphaGo tries to maximize
its probability of winning, but it doesn't care at all about
the margin by which it wins.through self-playing,
reinforcement learning, it plays against different versions
of itself many millions of times,It's unbelievably superhuman, in general, but it has some particular weaknesses
in some situations.through self-playing, reinforcement learning, it plays against different versions of itselfcame up with something new, and creative, and different