Many Faces uses 2200 for RAVE_EQUIV.  I found that anything between 2000 and 
3000 was about the same, and CLOP recommended 2200.  1000 was a little worse, 
and 500 was much worse.  In discussions with other programmers I heard numbers 
between 1111 and 5000.

 

For parameter tuning I recommend Emi’s CLOP.

 

David

 

From: Computer-go [mailto:[email protected]] On Behalf Of 
Urban Hafner
Sent: Monday, October 26, 2015 8:38 AM
To: [email protected]
Subject: Re: [Computer-go] Understanding and implementing RAVE

 

On Mon, Oct 26, 2015 at 1:12 PM, Petr Baudis <[email protected]> wrote:

 

This RAVE formula has been derived by David Silver as follows
(miraculously hosted by Hiroshi-san):

        http://www.yss-aya.com/rave.pdf

Also note that RAVE_EQUIV (q_{ur} * (1-q_{ur}) / b_r^2) varies widely
among programs, IIRC; 3500 might be on the higher end of the spectrum,
it makes the transition from AMAF to true winrate fairly slow.  You
typically set the value by parameter tuning.

 

Thanks for the link. That’s actually understandable to me. :) It seems that 
very soon there will be no easy wins anymore and I will have to invest the 
resources into some parameter tuning.

 

Urban

-- 

Blog: http://bettong.net/

Twitter: https://twitter.com/ujh

Homepage: http://www.urbanhafner.com/

_______________________________________________
Computer-go mailing list
[email protected]
http://computer-go.org/mailman/listinfo/computer-go

Reply via email to