Author: Rolf Tueschen
Date: 10:32:58 01/05/04
Go up one level in this thread
On January 05, 2004 at 03:14:48, Uri Blass wrote: >On January 04, 2004 at 23:04:55, Christophe Theron wrote: > >>On January 04, 2004 at 11:10:41, Bob Durrett wrote: >> >>>On January 04, 2004 at 11:00:31, Dan Andersson wrote: >>> >>>> I admire your persistance. I guess most of us that have a mathematical >>>>statistics education got tired explaining things after the first thread or so. >>>> >>>>MvH Dan Andersson >>> >>>I, too, have a "mathematical statistics education." >>> >>>What bugs me is that all of the CCC bulletins seem to suggest that those who run >>>and evaluate tournaments look only at the win/loss statistics. There is >>>considerably more information in a game score than just the final game result. >>> >>>Throwing away useful information is what I call "blind adherence to statistics." >>> One needs to rise above one's formal education and supplement it with good >>>thinking. >>> >>>: ) >>> >>>Bob D. >> >> >> >>The games themselves do not contain more information about the relative strength >>of the opponents than the bare winning percentage of the winner. >> >>That should not be forgotten. >> >> >> >> Christophe > >No > >The games have more information but the problem is how to interpret them. >Let give an extreme example when it is easy to learn from only 2 games who is >better. > >Suppose that the loser program in both games does mistakes that 2 ply search can >avoid and not one mistake. > >First it is losing a pawn and later it is losing a knight and later the queen >and finally it is checkmated. > >Suppose that it happens in 2 games. > >You can say that the winner is better based only on the games but you cannot say >it based only on the results. > >Usually learning from the games is harder but it does not mean that they have >not more information then the results. > >Uri Uri, there is a triviality we must understand: IF we had a one-dimensional complexity then we could take a single game and the result would tell us all. When we have a result and say that it is combined with a 90% significance we touch the multitude of dimensions of the whole complexity. The differences in strength (of machines) are of course related with chess and its dimensions. With our statistical universe we are trying to paint a mathematical mirror of chess. If you have two different versions of your program and want to know which one is stronger, you should always work with a continual fixation on just one changed factor. If you don't and - say - take two factors who are both changing you might overlook that the one is a good invention while the other is blocking the other only. Then the pure chess analysis might help you. Statistics alone brings you zero. That is also the reason why I must smile when I read that many programmers are very weak chessplayers. They _must_ work on the 1-variable-floating design. I think the first who used that method was Ed Schroder with his autoplayer kitchen and Ed isn't a bad chessplayer himself. But he's not a master. So he relied on the overnight results for the actual tunes. Bob Hyatt told us many anecdotes about his cooperation with different GM who "tested" Crafty in series of games. Of course a GM then could tell Bob that - say - the far distant pawns should get a higher eval overall. But it is all a question of the ability to see perspectives at a possibly earlier time. IMO - just because R. LANG was a lousy chessplayer and still for a long time number one commercial player it is wrong to deny that computerchess can only improve with chess in the first most important dimension. The actual hardware dizziness=delusions hides that. The book "cheat" [Rolf T.] hides even more structural stupidiness. It is painful but we should understand that a veritable GM knows what is stupid with the actual machines! In the so-called show events he is being paid to keep silence about the weaknesses. Weaknesses, weaker chessplayers couldn't exploit because they have no developped perception for such aspects. Rolf
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.