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

BGonline.org Forums

Live variance reduction tournament

Posted By: Timothy Chow
Date: Wednesday, 7 September 2011, at 5:27 p.m.

In Response To: Live variance reduction tournament (misja)

misja wrote:

Perhaps using the difference between the two error rates as the criterion would do the same as using the luck adjusted result?

They're not quite the same. (First of all, it's important to use MWC rather than PR or ER to measure error, in order to get good mathematical properties. I'll assume that we're using MWC in what follows.) In the long run, the luck-adjusted result will avoid any bias caused by the bot being wrong in its assessments, whereas the error-difference method will be biased. On the other hand, the error-difference method should be much less noisy than the luck-adjusted result in the short run.

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.