Author: ShaktiFire
Date: 13:45:19 07/15/00
Go up one level in this thread
On July 15, 2000 at 16:31:43, Dann Corbit wrote: > >I think that there may be some broad confusion as to even what the GM-Comp >strength arguements are about. > >Nowhere in my posts or in most of the other "it's not decided" proponents will >you see the statment "Computers are not GM strength." > >What I am arguing over is jumping to conclusions. The reason I make such a big >fuss over it is because science and mathematics are often incredibly >counter-intuitive. > >Consider a disease. We'll call it 'fooble' because the real name does not >matter. Now, we find a chemical which we believe will cure fooble. So we get >10,000 fooble patients and give them the pill. Sure enough 4 of them get >better! You might think that because it definitely happened (and more than once >-- four times!) that this was evidence that the chemical helps. But it turns >out that of the group given the plain sugar pill, also 4 people were cured. >Sound unrealistic? It happens all the time. > >When we make an observation that matches what we *wish* to be true, it is very >much human nature to assume that this observation proves the hypothesis. But >such an observation is not a proof at all. Confirmation of a hypothesis is >something that must be done statistically in order to be sure that the outcome >is really true and not just what we wish it to be. > >Many persons might think that it will be some kind of victory against persons >such as myself or Dr. Hyatt when computers are proven to be of GM strength. (And >I am certain that at some point they will be). Actually, it will be a victory >for science, and hence both for myself and for Dr. Hyatt also. > >I believe (and perhaps he will correct me if I am wrong) that Dr. Hyatt has not >said anywhere that Deep Junior is not of GM strength. He has only said that it >has not been adequately demonstrated. > >I think he is still leaning towards 'they might not be that strong' while I am >leaning towards 'they probably are that strong' but in either case, the >arguments you see are not about the conclusions themselves, but about jumping to >the conclusions and assuming that there is adequate proof to support those >conclusions. > >At this stage, the hypothesis has not yet gained theorem status. That does not >mean the hypothesis is incorrect. Only that it has not been adequately >demonstrated. > >In math, there is a latin abbreviation Q.E.D. which means basically when >translated, "The thing which was to be proven has been adequately demonstrated." > >We're not there yet. > >Those that say, "But look at the proof!" are simply unaware that such things do >have statistical measures. These measures can be firmly founded, and used to >demonstrate the thing which is to be proven. > >To be of GM strength does not mean "To have a scintillating performance." Many >people have scintillating performances but are not of GM strength. To be made a >GM, one must have broad excellence over a long period of time and also to have >several scintillating performances. > >Is Deep Junior a GM (Or even a super GM?) > >The best answer is "Maybe." Yes. But to be precise, you have to ask what the definition of GM is. If it is a title, only granted by performing to a certain standard in tournament play, sanctioned by FIDE, etc. Then Deep Junior is not a GM. The question is: Do computers play at GM level in standard time play? Your answer: maybe is a correct answer. However, your view that there is not enough information to answer the question is dubious. Chris Carson has documented dozens of games at standard time control of computer play vs. GMs. I won't knit pick...this or that program, this or that hardware. But in the last 2 years, dozens of games have been played. Computers vs. GMs at standard time control. Ratings can be calculated with these games. The more games played, the less uncertainty in the rating. The rating indicated, based on these dozens of games is over 2500. Does over 2500 rating, based many games, indicate "GM level of play". Maybe.
This page took 0.02 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.