Dan Andersson wrote:
No more incredible than that Mini-Max and Alpha-Beta will generate perfect play given enough resources. In the worst case MC/UCT will build a Mini-Max tree solving the game.
I've not looked very well at how UCT works, but surely this depends on the randomness of the MC? What if by some bizarre coincidence the source of random bits gives all 1's? What if it's pseudo-random, and interferes with itself so some paths are never investigated?
Or are you stipulating not just a good random source, but one which sooner or later enumerates all possibilities?
_______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
