On Mon, May 25, 2009 at 11:22 AM, Stefan Kaitschick < stefan.kaitsch...@hamburg.de> wrote:
> Almost any strategy other than playing out all legal moves >> involves a lot of hand waving that is unlikely to be >> accepted as a proof. There are just too many cases where >> a pitch inside a captured space has global effects. >> > > Completely solving small goboards seems like a strange quest to me. > Certainly there is no hope of scaling this upwards, so what's the point? > All of these things, including solving the game on small boards is a valid endeavor for it's own sake and almost always something useful will come out of it that can be applied elsewhere in other ways. Such a solver would be useful for learning how to generate heuristics to play the larger game better perhaps. But even if it wasn't, it's a fun puzzle on it's own merit. A program to play the big game is just as frivolous in the grand scheme of things, so what's the point of that? What is the point of solving a crossword puzzle? Where is the imagination to make such a statement? > > When nine men's morris was solved, I thought this was interesting, but tiny > go boards just don't make my blood boil. > But what I do find interesting about this, is the question of terminating > playouts. This is relevant for large boards also. This is actually a problem and it's interesting to me also. You can play to the end but is there a fail-proof way to handle seki correctly? > > And somehow I don't ever see comments anywhere suggesting that this could > be a problem. So what I'd like to know is: is this so trivial that no one > ever mentions it, or are the heuristics that programs use to terminate > playouts so obscure that they are too embarrasing to mention? I think it's a problem. I think 99.99 percent of all seki could be covered by simple rules. There is also the eye-filling question, is there a perfect rule for saying when an eye should not be filled? I wanted to build a 5x5 solver, starting with smaller boards first, but I was stumped by these problems. There is a way around the eye issue and probably the seki issue too. Just define the problem away. Build a solver and state clearly that this solver plays with a special rule that it is illegal to move into a 1 pt eye (and then define what a 1 point eye is.) Your disclaimer is that you make no claims about other rulesets. I don't think that is so bad, since every ruleset in GO, and there are many of them, invalidates results and conclusions drawn from another set. With seki, it is difficult to know exactly when to stop, you cannot just look at a given board and always say that this is a terminal node. - Don > > > Stefan Kaitschick > > > _______________________________________________ > computer-go mailing list > computer-go@computer-go.org > http://www.computer-go.org/mailman/listinfo/computer-go/ >
_______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
