________________________________ From: Don Dailey <[email protected]> On Thu, Jun 2, 2011 at 9:00 AM, Jacques Basaldúa <[email protected]> wrote: >If there is a skillful move that requires a precise sequence and >the playouts get it wrong, the tree will not explore that direction >frequently enough to find the skillful sequence by itself. It may >take a million years and more RAM that is available on the planet. > > But that has nothing to do with anything relevant here. The fact of the >matter is that > even the very best MC programs suffer from positions where a million years of >computing will still give the wrong results. No, that's not exactly Hideki's point. The position itself isn't inherently difficult - a double-digit-kyu player can usually get it right. The problem is that, given a particular (flawed) playout policy, the solution cannot be computed in any reasonable amount of time. The flawed policy explores the wrong part of the tree, in a systematically biased way. It's like probing a haystack with random pokes of a magnetic pole, not knowing that there is an anti-needle bias built into the pole, which avoids those particular spots in the haystack.
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
