[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnubg] Measuring performance levels 2
From: |
Joseph Heled |
Subject: |
Re: [Bug-gnubg] Measuring performance levels 2 |
Date: |
Wed, 30 Oct 2002 12:08:28 +1300 |
Hi,
We seriously need to consider changing the move selection pruning to something
more configurable but with good defaults. Here is my suggestion.
For each ply N, you can specify the amount of filtering of 0,1, ..., N-1 plys.
Each filtering level is given by M, T and M(T). M is the maximum number of moves
to keep. M(T) is the number of additional moves to keep, assuming they are more
than T from the best move.
(I am talking equity here, assuming usage of 'gammon weights')
So, for example,
Ply 1: 8,0,0 -- evaluate top 8 0ply moves by 1ply.
Ply 2: 2,3,0.1 -1,0,0 -- evaluate between 2 to 5 moves as given by 0ply.
Don't use 1ply.
Ply 3: not sure, I hardly use that
Comments and volunteers?
-Joseph
Morten Wang wrote:
>
> * Joseph Heled
> > Can you see if you have disabled 1 ply pruning? (i.e. enabled "No 1-ply
> > pruning")
>
> 1-ply pruning was not disabled (it doesn't seem to be by default with
> neither "world class" nor "world class++" settings.
>
> > After you do so, can you run the rollout again? Just start with a
> > low number of trials and see if it made a difference.
>
> 160 trials with the same setting and no 1-ply pruning gives:
>
> Win (g) (bg) L(g) L(bg) Cubeless
> 0.2170 0.0000 0.0000 0.0000 0.0000 -0.5657
> 0.0127 0.0000 0.0000 0.0017 0.0001 0.0255
>
> Morten!
>
> --
> "...a liquid which was almost, but not quite, entirely unlike tea."
> My opinions are not necessarily those of my provider,
> not necessarily mine, and probably not necessary.
>
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/bug-gnubg
- Re: [Bug-gnubg] Measuring performance levels 2, (continued)
- Re: [Bug-gnubg] Measuring performance levels 2, Joern Thyssen, 2002/10/26
- Re: [Bug-gnubg] Measuring performance levels 2, Joern Thyssen, 2002/10/26
- Re: [Bug-gnubg] Measuring performance levels 2, Douglas Zare, 2002/10/28
- Re: [Bug-gnubg] Measuring performance levels 2, Morten Wang, 2002/10/28
- Re: [Bug-gnubg] Measuring performance levels 2, Joern Thyssen, 2002/10/28
- Re: [Bug-gnubg] Measuring performance levels 2, Morten Wang, 2002/10/28
- Re: [Bug-gnubg] Measuring performance levels 2, Joseph Heled, 2002/10/29
- Re: [Bug-gnubg] Measuring performance levels 2, Joseph Heled, 2002/10/29
- Re: [Bug-gnubg] Measuring performance levels 2, Morten Wang, 2002/10/29
- Re: [Bug-gnubg] Measuring performance levels 2, Joseph Heled, 2002/10/29
- Re: [Bug-gnubg] Measuring performance levels 2,
Joseph Heled <=
- Re: [Bug-gnubg] Measuring performance levels 2, Jim Segrave, 2002/10/30