This sounds just like Lambda search.
On 6/19/05, Peter Drake <drake@xxxxxxxxxxxxxxxxx> wrote:
In tactical search, the defender is trying to build a certain
structure (roughly, two eyes) and the attacker is trying to prevent
this. I believe other placement games like Hex and Go-Moku also have
this property, but Chess (largely) does not.
This constructive property of the game suggests an interesting search
strategy: first find out how I could accomplish my goal if the
opponent kept passing, then consider potential countermoves and try
to "repair" the plan. This would allow us to abandon some searches
very early ("I can't make two eyes no matter how many moves I get!")
and could greatly narrow certain others ("If I get to make these
three moves, I'm alive, so you have to stop that.").
Does anyone know of work on this? Obvious fatal flaws?
Peter Drake
Assistant Professor of Computer Science
Lewis & Clark College
http://www.lclark.edu/~drake/
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/