Re: [Bug-gnubg] GNUBG gammon bug

[Jim Segrave]

On Sun 20 Jul 2003 (12:52 -0400), address@hidden wrote:
> Hi,
> 
> I analysed a match my nephew had played on FIBS with GNUBG 0.14 (build
> Jul 2 2003), analysis setting was supremo for checkerplay.
> In the situation shown below GNUBG critised Spock's move as a huge blunder, 
> but I can't see why; to the contrary, isn't GNUBG's evaluation extremely 
> faulty 
> ?
> Thanks for any explanations !
> 
> melBOT (O, 1 pts) vs. Spock (X, 0 pts) (Match to 5)
> 
> Game number 2
> 
> Move number 72: X to play 33
> 
>  GNU Backgammon  Position ID: FgAAwJ8dRAACAA
>                  Match ID   : QYmtABAAAAAA
>  +13-14-15-16-17-18------19-20-21-22-23-24-+     O: melBOT (Cube: 2)
>  |    X             |   |          O  O  X | OOO 1 point
>  |                  |   |             O    | OOO 
>  |                  |   |                  | OO  
>  |                  |   |                  | OO  
>  |                  |   |                  | OO 
> v|                  |BAR|                  |     5 point match
>  |                  |   |          7       |    
>  |                  |   |          X       |     
>  |                  |   | X        X       |     
>  |                  |   | X  X     X       |     Rolled 33
>  |    X             |   | X  X     X       |     0 points
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+     X: Spock
> Pip counts: O 7, X 98
> 
> * Spock moves 14/5 11/8
> Alert: very bad move (-9,664%)
> 
> Rolled 33 (+0,074%):
>      1. Cubeful 2-ply    11/8 6/3(3)                  MWC:  14,85%
>        0,000 0,000 0,000 - 1,000 0,419 0,235
>      2. Cubeful 2-ply    14/11 6/3(3)                 MWC:  11,51% ( -3,35%)
>        0,001 0,000 0,000 - 0,999 0,550 0,358
> *    3. Cubeful 2-ply    14/5 11/8                    MWC:   5,19% ( -9,66%)
>        0,002 0,000 0,000 - 0,998 0,799 0,589
>      4. Cubeful 0-ply    14/5 6/3                     MWC:   3,83% (-11,03%)
>        0,007 0,000 0,000 - 0,993 0,865 0,209
>      5. Cubeful 0-ply    11/5 6/3(2)                  MWC:   3,76% (-11,09%)
>        0,004 0,000 0,000 - 0,996 0,861 0,173

I'm very glad you raised this, it caused me to find and fix a rather
stupid bug in the new rollout code. Having fixed it, the answer is:

Yes, I'd say this is a position the neural net evaluates poorly. A
rollout (dropping moves more than 1.96 jsd's from the best) says that
your nephew's move is the best one in a very dismal situation:

gnubg (O, 1 pts) vs. jes (X, 0 pts) (Match to 5)


    GNU Backgammon  Position ID: FgAAwJ8dRAACAA
                    Match ID   : QYmtABAAAAAA
    +24-23-22-21-20-19------18-17-16-15-14-13-+  O: gnubg (Cube: 2)
OOO | X  O  O          |   |             X    |  1 point
OOO |    O             |   |                  |  
 OO |                  |   |                  |  
 OO |                  |   |                  |  
 OO |                  |   |                  |  
    |                  |BAR|                  |v 5 point match
    |       7          |   |                  |  
    |       X          |   |                  |  
    |       X        X |   |                  |  
    |       X     X  X |   |                  |  Rolled 33
    |       X     X  X |   |             X    |  0 points
    +-1--2--3--4--5--6-------7--8--9-10-11-12-+  X: jes
Pip counts: O 7, X 98

* jes moves 14/5 11/8
Alert: very bad move (+0.000%)

Rolled 33 (+0.080%):
*    1. Rollout          14/5 11/8                    MWC:   3.40%
         0.0%   0.0%   0.0% - 100.0%  86.6%  72.6% CL   3.43% CF   3.40%
      [  1.6%   0.0%   0.0% -   1.6%   0.5%   0.7% CL   0.14% CF   0.14%]
        Full cubeful rollout with var.redn.
        1296 games, Mersenne Twister dice gen. with seed 1 and quasi-random dice
        Play: 0-ply cubeful [expert]
        Cube: 0-ply cubeful [expert]
     2. Rollout          14/2                         MWC:   2.95% ( -0.45%)
         0.2%   0.0%   0.0% -  99.8%  88.6%  73.7% CL   2.99% CF   2.95%
      [  0.0%   0.0%   0.0% -   0.0%   0.5%   0.8% CL   0.12% CF   0.13%]
        Full cubeful rollout with var.redn.
        893 games, Mersenne Twister dice gen. with seed 1 and quasi-random dice
        Play: 0-ply cubeful [expert]
        Cube: 0-ply cubeful [expert]
     3. Rollout          14/5 6/3                     MWC:   2.72% ( -0.68%)
         0.0%   0.0%   0.0% - 100.0%  89.2%  74.0% CL   2.73% CF   2.72%
      [  0.9%   0.0%   0.0% -   0.9%   0.9%   1.1% CL   0.24% CF   0.24%]
        Full cubeful rollout with var.redn.
        773 games, Mersenne Twister dice gen. with seed 1 and quasi-random dice
        Play: 0-ply cubeful [expert]
        Cube: 0-ply cubeful [expert]
     4. Rollout          14/8 5/2(2)                  MWC:   2.71% ( -0.69%)
         0.0%   0.0%   0.0% - 100.0%  89.3%  75.1% CL   2.72% CF   2.71%
      [  0.0%   0.0%   0.0% -   0.0%   0.5%   1.0% CL   0.12% CF   0.12%]
        Full cubeful rollout with var.redn.
        608 games, Mersenne Twister dice gen. with seed 1 and quasi-random dice
        Play: 0-ply cubeful [expert]
        Cube: 0-ply cubeful [expert]
     5. Rollout          11/2 6/3                     MWC:   2.52% ( -0.88%)
         0.2%   0.0%   0.0% -  99.8%  90.2%  75.8% CL   2.56% CF   2.52%
      [  0.2%   0.0%   0.0% -   0.2%   0.9%   1.5% CL   0.25% CF   0.26%]
        Full cubeful rollout with var.redn.
        411 games, Mersenne Twister dice gen. with seed 1 and quasi-random dice
        Play: 0-ply cubeful [expert]
        Cube: 0-ply cubeful [expert]

Output generated Mon Jul 21 00:51:56 2003
by GNU Backgammon 0.14-devel (Text Export version 1.48)



-- 
Jim Segrave           address@hidden