Author: Enrique Irazoqui
Date: 03:34:24 05/25/00
Go up one level in this thread
On May 25, 2000 at 06:11:50, Jouni Uski wrote: >On May 25, 2000 at 05:57:44, Enrique Irazoqui wrote: > >>On May 24, 2000 at 18:00:12, Robert Hyatt wrote: >> >>>On May 24, 2000 at 15:08:15, Ernst A. Heinz wrote: >>> >>>>Dear Fellow Computer-Chess Enthusiasts, >>>> >>>>In view of the current discussion about diminishing returns in the thread >>>>"Ply Depth in Relation to ELO again", I like to share the results of my >>>>latest self-play experiment with you. >>>> >>>>The stunning outcome of the new experiment is that it shows the existence >>>>of diminishing returns for additional search in computer chess self-play >>>>with 95% statistical confidence, exemplified by the program "Fritz 6"! >>>> >>>>The title and abstract of my M.I.T. LCS Technical Report on the >>>>experiment follow below. >>>> >>>>*********************************************************************** >>>> >>>> ``A New Self-Play Experiment in Computer Chess'' >>>> >>>> ABSTRACT >>>> >>>>This paper presents the results of a new self-play experiment in >>>>computer chess. It is the first such experiment ever to feature search >>>>depths beyond 9 plies and thousands of games for every single match. >>>>Overall, we executed 17,150 self-play games (1,050--3,000 per match) >>>>in one "calibration" match and seven "depth X+1 <=> X" handicap >>>>matches at fixed iteration depths ranging from 5--12 plies. For >>>>the experiment to be realistic and independently repeatable, we relied >>>>on a state-of-the-art commercial contestant: "Fritz6", one of the >>>>strongest modern chess programs available. The main result of our new >>>>experiment is that it shows the existence of diminishing returns for >>>>additional search in computer chess self-play with 95% statistical >>>>confidence, exemplified by the program "Fritz6". The diminishing >>>>returns manifest themselves by declining rates of won games and >>>>reversely increasing rates of drawn games for the deeper searching >>>>program versions. The rate of lost games, however, remains quite >>>>steady for the whole depth range of 5--12 plies. >>>> >>>>*********************************************************************** >>>> >>>>Please find the full report in gzip'ed PostScript format at the URL >>>><http://supertech.lcs.mit.edu/~heinz/ps/new_exp.ps.gz>. >>>> >>>>Any comments welcome! >>>> >>>>=Ernst= >>>> >>>>P.S. >>>> >>>>Electronic preprints of my earlier publications on the relationship >>>>between computing power and playing strength of chess programs are >>>>available from http://supertech.lcs.mit.edu/~heinz/ and the WWW >>>>pages of "DarkThought" at http://supertech.lcs.mit.edu/~heinz/dt/. >>> >>> >>>The idea is ok, but I don't like the concept of playing program X vs itself >>>with different depths. Your conclusion can easily be right for Fritz, but >>>wrong for other programs... It would be hard to draw conclusions based on >>>testing only one program that is known to be very fast but not very 'smart'. >> >>In my opinion, this thing about Fritz being "fast and dumb" is a common place >>with little grounds. Fritz is certainly fast, but not more dumb than anything >>else I know, including "smart" programs like Hiarcs, Rebel and Mchess. Since >>these programs don't take advantage of their "knowledge" when playing the so >>called "fast and dumb", we should start questioning their smartness. Same for >>the way they all fall for anti-computer traps. Rather than fast versus clever, >>what I see is variations on the same theme. >> >>Enrique > >I agree. Actually Fritz is no more so fast. E.g. LG2000 is much faster and even >Junior is a little faster! Do you mean faster in NPS? Because tactically Fritz 6a is the fastest of all. So let's see: Fast and dumb versus slower and dumb versus slooooow and dumb, etc. Common denominator: dumb. On the other hand, humans are very dumb for not realizing in a split second that 8/8/8/1p6/4P3/2K1P3/k7/8 b - - 0 1 is a mate in 128. I mean, dumb for a machine may not mean the same as dumb for a human. Enrique >Jouni
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.