A machine learning algorithm, Alpha Zero, recently defeated the leading chess computer and revolutionized centuries of chess theory. After studying the game for four hours.
-
Show this thread
-
Replying to @Vermeullarmine
A chess engine or a champion centaur? What else had the algorithm studied? Link?
1 reply 0 retweets 0 likes -
Replying to @NGrossman81
No human input. It played itself for four hours, that’s all.
2 replies 0 retweets 1 like -
-
Replying to @NGrossman81
Stockfish! Which had just won the computer world championship.
1 reply 0 retweets 2 likes -
Replying to @Vermeullarmine
Holy crap. And I just realized it's the same underlying engine as AlphaGo.
2 replies 0 retweets 2 likes -
Replying to @NGrossman81
I'd be curious to see the same algorithm applied to unbounded strategy. The number of possible positions on a chess board is in the range of 10^45. That means a sufficiently powerful computer could plausibly play every game...
2 replies 0 retweets 0 likes
But something with known and unknown unknowns, like multiplayer no-limit poker, would task it in a different way. Brute force, no matter how creative, wouldn't cut it alone.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.