It's then the algorithm that is responsible, and absolutely not the
specifics of a game.
Exactly, it is the algorithm. I think we agree on all points.
Not quite all points yet. The calculations in my penultimate message
still apply to UCT. So we're back to finding a reason why speculative
pondering would be inefficient. My first guess is decreasing returns.
It just happens that the most successful search algorithm for chess
makes speculative pondering work, whereas in Go it is the opposite.
Though it's true that it's much more obvious when we have to add
a P[match] factor also for usual pondering, as would be the case for
alpha-beta.
Jonas
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
