Author: Robert Hyatt
Date: 19:29:56 07/04/01
Go up one level in this thread
On July 03, 2001 at 10:48:27, Peter Berger wrote: >On July 03, 2001 at 10:40:33, Christoph Fieberg wrote: >> >>After this surprising result I investigated the Kasparov-Kramnik match. What is >>the likelihood that a 2849 rated player does not win a single game against a >>2770 rated player and even is defeated 2-0 after 15 games? >> >>The outcome was shocking! The chance is only 1:1439 (likelihood 0.0695%!). > >I think for a match beween humans your model won't work at all. > >It can be argued if the results of the single games are statistically >independent in a comp-comp match ( I think they are still independent enough >that your model should work well ). > >But in a human match they definitely aren't. > >pete > I don't think they are independent events with comp vs comp either. Book learning. Position learning. Both "connect" a series of games together. Even more so if one program has those and the other does not. >> >>This results changes my view on the outcome of the Fritz-Junior match. >> >>Nevertheless it is really astonishing what happened in these matches. >> >>Christoph
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.