Author: James T. Walker
Date: 17:48:44 01/11/01
Go up one level in this thread
On January 11, 2001 at 16:41:45, Dann Corbit wrote: >On January 11, 2001 at 16:28:04, Dann Corbit wrote: > >>On January 11, 2001 at 16:06:10, James T. Walker wrote: >> >>>On January 11, 2001 at 16:01:16, Dann Corbit wrote: >>> >>>>On January 11, 2001 at 15:58:55, James T. Walker wrote: >>>> >>>>>On January 11, 2001 at 14:59:06, Dann Corbit wrote: >>>>> >>>>>>On January 11, 2001 at 14:46:01, Garry Evans wrote: >>>>>> >>>>>>>On January 11, 2001 at 13:39:03, Dann Corbit wrote: >>>>>>> >>>>>>>>On January 11, 2001 at 11:43:10, Drazen Marovic wrote: >>>>>>>> >>>>>>>>> The sad thing is, if rebel had lost by a measly half point countless here >>>>>>>>>would still try to deny comps gm strength. >>>>>>>> >>>>>>>>There is not enough evidence to confirm or deny the assertion either way. >>>>>>>> >>>>>>>>The 1/2 point swing in the other direction (for the comp) is no different. But >>>>>>>>in any case, there are certainly not enough games to make a logical statement. >>>>>>>>Only an emotional one. >>>>>>> >>>>>>>Baloney! We have more than enough games, simply visit Chris Carson's chess page. >>>>>> >>>>>>Been there, done that. >>>>>>Take the individual combinations of machine and program, and calculate the error >>>>>>bars for ELO. They are close to infinity. >>>>> >>>>> >>>>>Why do that? Why not assume that all computers are essentially equal and go >>>>>from there? If only one is of GM strength then using all of them could only >>>>>decrease the argument for GM strength. >>>> >>>>We can do anything we like and make any sort of assumptions that we choose. >>>>Fortunately, we have mathematics to test our models afterwards. Without using >>>>this tool, we are making emotional choices rather than logical ones. There is >>>>nothing wrong with that, but (personally) I prefer a rational approach. >>> >>>Please explain, mathmatically if you like, what is wrong with my proposal. What >>>part is irrational/emotional? >> >>Assume that all computers are essentially equal? >>;-) >> >>OK. You get a Commodore 64 and I get the latest Cray. Ready for a game of >>chess? > >I'm not altogether clear here, let me add to my remark. > >If we start by assuming something we know is not true (e.g. all computers are >equal in ability) then we have added a false axiom to our system. The inclusion >of a single false axiom means that every false statement can be proven to be >true. Therefore, such a step is inadmissable. > >However, we can start with a similar statement: >A program will not vary in strength by changing machines by more than 100 ELO. >Now *this* hypothesis, on the other hand, is verifiable and testable. If this >is demonstrated to be correct, we can use it as an assumption for all the >systems on which we have demonstrated it. (BTW, I think it is probably false). > >Consider also: >A computer program will have bugs. All of them do. Given enough games, GM's >will find a blind side and exploit it (if they have the interest to do it). >Even computers can do this. In fact, that is why (in my opinion) the strongest >programs are _not_ at the top of the SSDF list. The programs which make the >best chess decisions [Rebel, Chess Tiger -- and this is pure conjecture on my >part I will admit] do not learn very well. Hence, learners will eventually >start to thump them. But if we erase the learning files and start over, I think >the tables would turn. > >Now, this *is* a defect in the programs that do not learn well. But as a >measure of a programs ability to analyze a position, I think the result could be >a faulty conclusion. You seem to be the one who is emotional and irrational now. Why go off on the deep end on something which is really simple. I was simply suggesting to take the games played by top programs in the last year or so and consider them all as one player. It is perfectly logical to assume that if only one program is of GM strength which many people claim is not, and you add the results of other programs to the statistics, you are taking a worst case scenario. This is true because the other programs surely are not GM strength if even 1 is not GM strength. This might give you enough games combined to determine the "average" strength of top programs today vs humans. Your main contention seems to be that there is not enough data to determine what the strength of Rebel is but you don't suggest how many games vs humans it would take to establish the fact one way or the other. How many games does it take for a human to establish himself/herself as equal to a GM in strength? What is GM strength? Maybe you can come up with a number which would satisfy most people or at least yourself. It's kind of like fuzzy logic. It becomes an easier and simpler way to arrive at the answer without demanding you og exactly where you want to go on the first try. It's obvious that computers will never hold a GM title because has made this much more difficult for computers than humans. So the only thing I know to do is to come up with some figures which most people agree is equal to a GM. If you can't do this then you may never agree that computers are at last equal to a GM even when computers are beating the pants off of GMs. So what I was suggesting was to take the last X number of games by computers vs GMs and treat them as one player. If this "Average" computer is of GM strength then seems to me we have some GM strength computers. If they don't measure up now then we have not proven that there are no GM computers but at least we prove that as a whole they are not there yet. Of course you would want to chose the best few computers which will give you enough games vs humans to establish yes or no. (Not a C64) Say if it takes 40 or 50 games to satisfy you that computers have reached Gm strength then use as many of the top computer vs human games you need to get the 40 or 50 games. So the bottom line is if you can't decide how many games it takes and what rating is equal to a GM then you will never answer the question. But if you can do that then maybe you can have the answer already. Or maybe you're not interested in the answer but just like to argue. Jim
This page took 0.01 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.