Re: [Bug-gnubg] Strange rollout result ?

[Neil Robins]

I think this is the sort of result you often get in situations where a player
is hit after bearing some checkers off. 2-ply frequently seems to massively
over-estimate that players chances and take huge passes.


Massimiliano Maini wrote:
> 
> Hi all,
> 
> this looks odd to me:
> 
>  GNU Backgammon  Position ID: cP8BAJBtGwCwBQ
>                  Match ID   : cAlgAXAACAAA
>  +13-14-15-16-17-18------19-20-21-22-23-24-+     O: gnubg
>  |                  | O | O  O  X  X  X    | O   7 points
>  |                  |   | O  O  X  X       | O 
>  |                  |   | O  O             | 
>  |                  |   | O                | 
>  |                  |   | 9                | 
> v|                  |BAR|                  |     11 point match (Cube: 1)
>  |                  |   |                  | 
>  |                  |   |                  | 
>  |                  |   |                  | 
>  |                  |   | X  X  X  X  X    |     On roll
>  |                  |   | X  X  X  X  X    |     1 point
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+     X: xxx
> 
> 
> Rollout details:
> Centered 1-cube:
>   0.8109 0.0000 0.0000 - 0.1891 0.0571 0.0050 CL  +0.5695 CF  +1.1167
>  [0.0019 0.0000 0.0000 - 0.0019 0.0014 0.0010 CL   0.0043 CF   0.0100]
> Player gnubg owns 2-cube:
>   0.8194 0.0000 0.0000 - 0.1806 0.0566 0.0071 CL  +1.1630 CF  +1.1385
>  [0.0018 0.0000 0.0000 - 0.0018 0.0015 0.0014 CL   0.0086 CF   0.0100]
> Full cubeful rollout with var.redn.
> 1296 games, Mersenne Twister dice gen. with seed 858145082 and 
> quasi-random dice
> Play:  2-ply cubeful prune [world class]
> keep the first 0 0-ply moves and up to 12 more moves within equity 0.16
> Skip pruning for 1-ply moves.
> Cube: 2-ply cubeful prune [world class]
> 
> In MWC:
> 
> Rollout details:
> Centered 1-cube:
>   0.8109 0.0000 0.0000 - 0.1891 0.0571 0.0050 CL  19.082% CF  21.356%
>  [0.0019 0.0000 0.0000 - 0.0019 0.0014 0.0010 CL   0.018% CF   0.041%]
> Player gnubg owns 2-cube:
>   0.8194 0.0000 0.0000 - 0.1806 0.0566 0.0071 CL  21.548% CF  21.446%
>  [0.0018 0.0000 0.0000 - 0.0018 0.0015 0.0014 CL   0.036% CF   0.042%]
> Full cubeful rollout with var.redn.
> 1296 games, Mersenne Twister dice gen. with seed 858145082 and 
> quasi-random dice
> Play:  2-ply cubeful prune [world class]
> keep the first 0 0-ply moves and up to 12 more moves within equity 0.16
> Skip pruning for 1-ply moves.
> Cube: 2-ply cubeful prune [world class]
> 
> 
> First, with ND at +1.1167 and D/T at 1.1385, gnubg indicates D/P as
> the right decision, while it should be ND/P (too good).
> 
> Second, there are no gammons for X, how can it be too good ?
> Is gnubg taking cobes that are in fact passes later on ?
> 
> By the way, the copy button in the hint window/analysis panel does not
> copy all the information (it misses the recap with the equities and
> differences dor the 3 actions ND, D/T, D/P and the final decision).
> 
> MaX.
> 
> P.S.
> I have a sgf file with the match and the rollout saved.
> 
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
> 
> 

-- 
View this message in context: 
http://www.nabble.com/Strange-rollout-result---tp23595836p23603478.html
Sent from the Gnu - Backgammon mailing list archive at Nabble.com.