bug-gnubg
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-gnubg] User training of the Neural Nets


From: Christian Anthon
Subject: Re: [Bug-gnubg] User training of the Neural Nets
Date: Fri, 25 Aug 2006 14:14:18 +0200

On 8/25/06, Ian Shaw <address@hidden> wrote:
TD-training to intermediate level (FIBS 1650).
Supervised training on rolled-out positions to get to advanced stage (FIBS 
1800). Positions included some where gnubg was known to err, and others chosen 
at random.
Supervised training on positions where 0-ply evaluation differed from 2-ply 
evaluation. Gnubg achieved expert level (FIBS 1930).


How was the last set of position chosen. As a subset of the previous?

Since Gnubg is now over the plateau reached by TD training, I wondered if a new 
bout of TD training on top of the supervised training might be beneficial. 
Øystein and Joseph, are you saying that you have already tried this, to no 
avail?

My guess is that it wouldn't work. As gnubg grew in strength, so did
the need to look at the problematic positions.

Christian.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]