[Top][All Lists]

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

Re: [Bug-gnubg] An evalutaion of the pruning nets

From: Nis Jorgensen
Subject: Re: [Bug-gnubg] An evalutaion of the pruning nets
Date: Tue, 26 Oct 2004 10:28:03 +0200
User-agent: Mozilla Thunderbird 0.8 (Windows/20040913)

Joseph Heled wrote:

>As the person who "independently discovered" the reduced evaluation (I am sure 
>others have thought about it before), I will be happy to see them go.
>Everything you can do with the reduced evaluation, you can do faster and with 
>better accuracy with the prune nets. The whole reason I developed the prune 
>was my dissatisfaction with reduced evaluations. I am not sure I have the time 
>to prove to you they are bad, but I am very confident that is the case.
>As to your idea of half ply, it is a good one. It certainly improves on 1ply 
>I certainly would love to see it in GNUbg. The way things work, this will 
>probably not happen before I implement it myself ...
Well, I actually implemented "fractional plies" myself - reusing both
the "reduced evaluation" code and the settings - so the only thing I
changed was the implementation of reduced eval. I don't remember the
exact history of why it wasn't committed.

The reason I modified the "reduced" code, and re-used the settings, was
that I think "fractional plies" (as I implemented them) is actually just
the correct way of implementing reduction, rather than a new concept
deserving it's own settings.

So unless someone disagree, I will try to get my fractional plies in
some time soon - so I would very much like for the "hooks" for reduced
evaluation (sgf-files, gtk-widgets etc)  to stay.



reply via email to

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