Author: Pete R.
Date: 09:15:43 07/26/00
Go up one level in this thread
On July 26, 2000 at 02:18:08, blass uri wrote: >On July 25, 2000 at 21:53:07, Pete R. wrote: > >>On July 25, 2000 at 18:57:44, Tom Kerrigan wrote: >> >>>On July 25, 2000 at 18:28:39, Dann Corbit wrote: >>> >>>>On July 25, 2000 at 18:23:05, Tom Kerrigan wrote: >>>>[snip] >>>>>Right. I think the only way to go is binary, i.e., "positive" instead of +0.22. >>>>>So the evaluation can be either right or wrong. >>>> >>>>I don't see any value in throwing away information. >>>> >>>>Your eval chooses ce=+1 >>>>Chess Tiger chooses ce = +340 >>>>You got it right? >>>> >>>>Your eval chooses ce = -20 >>>>Rebel chooses ce = -32765 >>>>You got it right? >>>> >>>>>Let's say you have a collection of 10,000 positions where you know which side is >>>>>winning. You run your evaluation function on these positions (which should only >>>>>take a few seconds) and get some output like: >>>>> >>>>>Eval function correct for 8,000 (80%) of the positions. >>>>> >>>>>Then you tweak the eval function and get 82%. You know your tweak was >>>>>beneficial. >>>> >>>>If your evaluation function got close to the right answer all the time, it was >>>>doing well. If it missed by 300% on average, it might be lame (see the above >>>>examples). >>>> >>>>I don't think that approach works. >>>> >>>>Centipawn evaluations are something of a crunchy continuum over the range of a >>>>short. To consider only what side of the zero it falls on (and what about >>>>zero?) is to throw away almost all of the information. How can you use this to >>>>create better decisions? >>> >>>Throwing away what information? Ideally, chess positions are evaluated to one of >>>three values. If you have any more information than that, you can be sure it's >>>flawed. >>> >>>If I aim to duplicate Crafty's evaluation function scores, then I will end up >>>recreating Crafty's evaluation function. That's not my intention. >> >>If that were simple to do then you could just use Fritz 6a or another top >>program, and match your evaluations. Your program would jump from amateur to >>top program in no time. > >evaluation is only part of the program. >If you have the same evaluation as Fritz6a but do not have the same search rules >you will not be in the same level. > >Uri That's what I'm saying, if it were so simple than any amateur program could quickly contend for the top of the SSDF list. But it might certainly be instructive to compare your evals to the top programs', and this is much easier than trying to find perfectly evaluated positions.
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.