I once read a very interesting paper about such an approach. It was
basically just describing a graph data structure where every group and
contiguous region of empty squares was represented by a vertex, and then
showing that various important properties of the position were preserved
even in this highly-compressed representation.

I'll try to find the paper online tonight, though I don't know how high my
chances are -- I originally saw it in a (paper) library.

Cheers,
Adrian

On Thu, Nov 18, 2010 at 3:25 PM, Steve Safarik <[email protected]>wrote:

>  Hi all.  I'm looking at some graph theory approaches to Go, and wondering
> if anyone has come up with a workable adjacency matrix representation for
> the board?  It's not entirely clear what "adjacent" means with regards
> to the interface between black & white stones, stones at a distance, etc.
> Same color stones next to each other is quite different from opposite color
> stones next to each other.  Any thoughts?
>
> Steve.
>
>
> _______________________________________________
> Computer-go mailing list
> [email protected]
> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
>
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go

Reply via email to