|
>Chrilly Donninger inquires about the lambda-algorithm. The algorithm is documented in the paper "Lambda-search in >game trees -- with application to Go" (from 2000) which is downloadable from here: www.t-t.dk/publications. The paper >comes in two 99.9% identical versions (lambda_lncs.pdf and lambda_icga.pdf). Read either, and there is no need to >look at the other. I have looked now at the code and the paper. I have seen it before in the ICGA-journal, but I skiped it at that time. "Yet another uninteresting Go article". But in fact the article is very interesting and well written.
In chess interesting mate-combinations have often a hierarchical inversion. Attacker sacrifices a piece with check, some more checks to drive king to a certain place, and then some "quiet" move (Lamda**2 or "loose net" in Go-terminology), possibly giving the defender the chance to make a revenge-check(s), defender runs out of counter threats and the final check-mating sequence starts. There are of course also combinations without the intermediate quiet moves. But these are very easy to solve. The quiet move can be a problem for Null-Move-Pruning. Actually it is a problem of the quiesence search. If the quiesence-search does not generate non-capturing checks such a combination can be very hard to find.
Chrilly
|
_______________________________________________ computer-go mailing list computer-go@xxxxxxxxxxxxxxxxx http://www.computer-go.org/mailman/listinfo/computer-go/