On Sun, 2008-02-10 at 18:35 +0100, Olivier Teytaud wrote: > > That translates to mean that MoGo no longer uses upper confidence > > bounds, and only uses means. It also means that MoGo will _never_ > > explore improbable children (after a few sims) unless the RAVE value > > yields an unusually high estimate for it. Is all of that correct? > > > > Precisely: I don't see why you would be wrong, but empirically for 9x9, > we have played games against high-level humans and for the (few :-) ) > games that mogo lost, we tried to see which moves were erroneously chosen > by mogo; if we restart mogo at the same position with a huge > computation time (30 minutes of a fast octocore) mogo always changed his > mind and moves to a better move.
I'm just surprised to hear that the program that introduced UCT (and got so many others to use it), isn't using UCT any more. Combining RAVE and UCT as described in the PDF still sounds like UCT to me, but with no sqrt(log) term, it no longer is. I'll certainly have to think about the trades being made and what I'd expect the outcome to be. 4 hours of CPU time seems like a really long time. Have there been any trades done to measure how long MoGo takes to change its mind under different configurations? > > So: > - theoretically, I don't see any reason for mogo to be asymptotically > consistent Asymptotically approaching perfect play is no longer a goal? (Rhetorical question to show some concern) > - there are long computation times during which mogo focuses on a bad > move > - however, we have not seen a case of bad move for which mogo keeps > this move in case of _very_ long computation times > > ==> if someone beats the release MoGoR3 with > very large computation times (time x nbcores = 4h, 1 to 4 cores) > I'm interested in the sgf file and the analysis I bet someone will take up the challenge. You should probably also give a limit to how long the human can think about the game. Can they also sit there for 4 hours contemplating a move and working out all the variations? Of course, this challenge by itself shows the ultimate effect of increased computational power... _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
