[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Are these related to your original issue?

Posted By: Chuck Bower
Date: Friday, 2 May 2008, at 7:37 p.m.

In Response To: a bias in rollouts (bob koca)

case 1) I analyze a match with SW (for example), using evaluations. Then I roll out only those plays where SW says I erred. Some of those positions actually get overturned (so I didn't err as many times as SW said). My error rate improves. Therefore I'm better than SW eval said I was. Therefore I make the general conclusion that humans aren't as bad as the bots evaluate them.

case 2) As above, and my rollouts only contain two candidates: the play I made and the play SW eval'ed as best. Is my final error rate result unbiased?

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.