Author: Chris Carson
Date: 16:42:53 04/17/02
Go up one level in this thread
On April 17, 2002 at 16:57:20, Roy Eassa wrote: >On April 17, 2002 at 16:36:03, Chris Carson wrote: > >>On April 17, 2002 at 15:59:18, Roy Eassa wrote: >> >>>On April 17, 2002 at 15:48:49, Roy Eassa wrote: >>> >>>> >>>>A given computer's rating will go down significantly (even though it does not >>>>gain or lose one iota of strength objectively) if and when its human opponents >>>>gain anti-computer skills. >>>> >>>>Does that make sense? >>>> >>>>I guess early ratings are one thing and ACTUAL STRENGTH is a different thing >>>>that is much harder to measure (requiring much more scientifically controlled >>>>circumstances). >>>> >>>>For humans versus humans, the two things (rating and actual strength) have >>>>tradionally been closely related, except when the player is a young child who is >>>>improving very rapidly. >>>> >>>>There is significant reason to believe that RATING and actual STRENGTH can get >>>>*way* out of sync with each other when it comes to computers, due to the extreme >>>>relevance of the anti-computer skills (and not normal chess skills) of the >>>>humans they have faced. >>> >>> >>> >>>Also, most (nearly all?) computers that have gotten an early rating (using fixed >>>hardware and software) have seen that rating drop SIGNIFICANTLY over time, as >>>humans learn better how to play well against computers. >>> >>>Does that mean: >>> >>>a) The computer is getting steadily weaker at chess? or >>> >>>b) Humans are quickly getting much better at chess? or >>> >>>c) A computer's early rating is NOT an accurate reflection of the computer's >>>actual chess strength, but is SKEWED by the fact that humans lack a special >>>skill that is required in order for them to score accurately against computers >>>-- a skill that is SEPARATE and distinct from the traditional skill most human >>>chess players have focused on? >>> >>>d) Some other explaination (please fill in)? >> >>If a GM only prepares for one other GM and his rating rises against that GM but >>falls with respect to the other GM's what does that say? > > >That's actually a slightly different topic. > >I was NOT actually referring to humans learning the weaknesses of a *specific* >program, but rather to the likelihood that most GMs will improve their overall >anti-computer abilities with time, and SOME of them will improve this ability a >LOT. > >Take a strong program running on a fast PC -- I'll agree with you that it could >get a 'rating' of 2700 in current GM matches -- and put the only copy of it in a >vault for 6 years. Let no human play it or study it during that time. Let the >best 3 anti-computer GMs of the year 2008 play matches against it. > >I think there's a very significant probability that that SAME program running on >that SAME hardware would then achieve a rating of 2500. Did it get 200 points >weaker sitting in the vault? > >I contend that the reason computers IN GENERAL have done as well as they have >against GMs in the past 2 years is that GMs have in general not learned a >necessary skill -- one that is SEPARATE from traditional chess skill -- that is >required ALONG WITH their chess skill in order to play computers. The reason >GMs have not learned this skill is simply that it has not been required until >very recently, whereas standard chess skills HAVE been required for centuries in >order to defeat humans. > >Interestingly, GMs have developed a whole set of psychological skills that ARE >required to do well against other human GMs but would NOT be required to do well >against computers. > >In conclusion, there are at least 3 separate skills involved: To score well >against humans, GMs need good chess skills AND good inter-human psychological >skills but NOT good anti-computer skills. To score well against computers, GMs >need good chess skills AND good anti-computer skills but NOT good inter-human >psychological skills. GMs _all_ have good chess skills and good inter-human >psychological skills. Most GMs do NOT YET have good anti-computer skills. > > > >> >>> >>> >>>Please choose one! I am not sure. You are suggesting a skill or specific knowledge. That is one thing. You also suggest that it is specific to only anti-computers. You also suggest that it will lower the rating of the computers significantly. This skill has been suggested and supported by individual experience (including my own). However, I would need proof in the form of hypothesis testing to ensure that skill can be quantified and tested and show significant influence. Test 1 - GM with skill can win significantly more than with out the skill against the computer. The anti-computer criteria would need to be quantified tested and show a significant influence for computers only. My guess is that it would also help against humans, but that can also be tested. Test 2 - Skill does not help GM win more against other GM's. As for ratings. You will have to to show that this skill is distributed enough amoung the rated population to significantly lower a programs rating. Thus with a random sample (so that rated players with the specific skill are not over or under represented) then the overall rating of the program is significantly lowered. Test 3 - This skill can be acquired by enough GM's to significantly lower the programs rating. This is interesting, I will not believe it until I see the results, however, it is worth investigating. Ofcourse this is just a discussion. The hypothesis, literature review and design of the three tests would be a large undertaking and must be done so that it can be repeated. I also think that by the time this is defined, quantified and tested it will be nulified by the programmers and a mute point.
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.