On Sat, Oct 30, 2010 at 6:56 PM, Jacques Basaldúa <[email protected]> wrote:
> About the convergence proof to go:
>
> As far as I can understand Álvaro's proof sounds correct,
> but that possibly applies to most games, but not go as
> humans play it. Because what we play is "MCTS go" an almost
> identical game where filling your own eyes is not allowed.

Yes, there is something tricky about this. As I stated the proof, you
would have to consider the game to be of depth 1 only when the
opponent has just passed and moving into any empty space would be a
super-ko violation. In this sense, the real depth of the tree is
something like 10^48: Simplifications (of the style of not filling
eyes in MC simulations) need to be made to get practical results.
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to