Author: Dann Corbit
Date: 14:37:19 06/02/04
Go up one level in this thread
On June 02, 2004 at 17:32:52, James Swafford wrote: >On June 02, 2004 at 17:25:55, Dann Corbit wrote: > >>On June 02, 2004 at 16:58:01, Dann Corbit wrote: >> >>>On June 02, 2004 at 16:07:14, James Swafford wrote: >>> >>>>On June 02, 2004 at 16:03:10, James Swafford wrote: >>>> >>>>>On June 02, 2004 at 10:06:22, Dann Corbit wrote: >>>>> >>>>>>On May 29, 2004 at 16:13:24, James Swafford wrote: >>>>>> >>>>>>>On May 29, 2004 at 14:15:37, Frank Phillips wrote: >>>>>>> >>>>>>>>On May 29, 2004 at 12:53:54, James Swafford wrote: >>>>>>>> >>>>>>>>>On May 29, 2004 at 11:53:29, Frank Phillips wrote: >>>>>>>>> >>>>>>>>>>On May 29, 2004 at 11:44:58, James Swafford wrote: >>>>>>>>>> >>>>>>>>>>>On May 29, 2004 at 11:35:07, Frank Phillips wrote: >>>>>>>>>>> >>>>>>>>>>>>On May 29, 2004 at 04:00:31, Gian-Carlo Pascutto wrote: >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>>>I don't think so. The program still has weaknesses that a bit of >>>>>>>>>>>>>extra hardware will not overcome. >>>>>>>>>>>>> >>>>>>>>>>>>>GCP >>>>>>>>>>>> >>>>>>>>>>>>What are these weaknesses? >>>>>>>>>>>> >>>>>>>>>>>>Bob may even be able to fix them before the event. >>>>>>>>>>> >>>>>>>>>>>He was talking about his program, not Crafty. >>>>>>>>>> >>>>>>>>>>Thanks. I misread the post. >>>>>>>>>> >>>>>>>>>>But I am still interested in the weaknesses being referred to by GCP, which are >>>>>>>>>>resistant to faster hardware. I have so many myself. If only I knew what they >>>>>>>>>>were :-) >>>>>>>>> >>>>>>>>> >>>>>>>>>As in, "I can't seem to mate Shredder, even with faster hardware!" ?? :) >>>>>>>>> >>>>>>>>>-- >>>>>>>>>James >>>>>>>> >>>>>>>> >>>>>>>>I guess the answer is yes, although I have never had better hardware - and am >>>>>>>>not SMP, so probably never will. >>>>>>>> >>>>>>>>See you tonight at ICC author's only tournament ? :-) >>>>>>> >>>>>>>NOt as a competitor-- my thing is nowhere near strong enough >>>>>>>to compete yet. I'm hoping to be able to compete in the next >>>>>>>CCT, though. >>>>>> >>>>>>Are you still doing the learning stuff? >>>>> >>>>>I've been working with TDLeaf quite a bit. At some point I'll >>>>>post something with some meat to it, but to sum it up, I'm >>>>>not nearly as optimistic about it as I once was. >>>>> >>>>>In my experience, TDLeaf can train the material weights, and it >>>>>can even produce an evaluation vector that's superior to a >>>>>'material only' vector. I am not convinced it's useful for >>>>>training a complex vector, nor am I convinced it does a better >>>>>job than hand tuning. For that matter, I am not even >>>>>convinced it converges to the optimal vector! >>>>> >>>>>Caveat: it's possible (though I think it's unlikely) that >>>>>my implementation is flawed. My engine will become open source >>>>>at some point (maybe after the next CCT), so you can judge >>>>>for yourself then. >>>>> >>>>>Will Singleton and I had a bet on this... I conceited defeat >>>> >>>> >>>>Gah! I "conceded" defeat. >>>> >>>>>the other day. THe original bet was for the loser to fly >>>>>the winner and spouse across country for drinks. :) I'm >>>>>pretty sure Will's decided he'll forego that if I show up >>>>>at a tourney, but that's his call. >>>>> >>>>>I'm still very interested in learning algorithms, but I'll >>>>>be focusing on improving my evaluation for a while. >>>>> >>>>>Again- I will post some data at some point. >>> >>>I am doing a computer guided optimization for Beowulf. >>> >>>It takes ~12,000 positions from super-GM games and SSDF games among the top >>>computers where all the participants chose the same move (no other moves chosen >>>for that position). >>> >>>For each of about 100 parameters, I vary the value from too small up to too >>>large (e.g. a knight might go from 200 centipawns to 450). At some optimal >>>point, the largest number of positions will be chosen. I fit a parabola >>>throught the data ans solve for the maxima (if any). >>> >>>Often, the variance of the parameter has no effect on the solution scores (for >>>instance, I might get 5500 solutions no matter what the parameter is, or the >>>number of solutions may vary randomly). So I also solve for the minima of the >>>time curve. As an example, a depth 4 search using NULL MOVE will probably solve >>>a few LESS positions than not using NULL MOVE, but it will take 1/3 of the time >>>at some optimal prune level. >>> >>>I have had lots of bugs in my curve analysis, but I am slowly working it out. >>> >>>Before, I solved for a smaller subset of tactical positions which made it great >>>at solving those tactical positions but lousy at playing. I am hoping for a >>>better result this time (especially since some of my result calculations were >>>backwards, making the fits enormously unstable). >> >>Here is the binary and source for the current project: >>ftp://cap.connx.com/pub/chess-engines/new-approach/beocurve.zip > >Alright: I'll take the bait... I'll download it and check it out. > >> >>There is one more correction in the file compared to my last runs -- It now >>compares the minumum of time fitted by the curve with the absolute minimum found >>in the raw data (before, that bit was wrong). >> >>Here is the curve for Bishop piece value: >> >>bishop_score=352 at 4; stddev=16.409341 : -0.0665458*x^2 + 46.9104*x + -2711.28 >>(x=291.000000, y=5298.000000), t=1137.000000 >>(x=307.000000, y=5431.000000), t=1141.000000 >>(x=323.000000, y=5504.000000), t=1143.000000 >>(x=339.000000, y=5519.000000), t=1145.000000 >>(x=355.000000, y=5566.000000), t=1145.000000 >>(x=371.000000, y=5539.000000), t=1146.000000 >>(x=387.000000, y=5473.000000), t=1145.000000 >>(xmax=355.000000, ymax=5566.000000), xmax seen verses curve xmax [355 352.968] >> >>I believe that the score will reduce at deeper plies (I have seen this trend at >>least for shallower plies so far). >>The result "bishop_score=352 at 4" means a biship_score of 352 centipawns is >>optimal for this test set at 4 plies deep searching. > >How long does it take to complete a test set at 4 ply? About an hour per parameter on a 2.4 GHz AMD 64 bit computer. >And- you only vary one parameter at a time, right? Yes. And I repeat a few cycles of the solution, because parameters have an effect on each other for obvious reasons. A 2 or 3 ply search is much faster. A 7 ply search will take eons. You can imagine the furious CPU activity even in a 4 ply search, as I am solving each of 12K Epd positions at 4 plies and at 6-12 different parameter values usually (on rare occasions they are binary or have only a few possible choices).
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.