If you change the number of simulation to 10,000 instead of 100,000 of
course you get more variety.  After only 200 runs I get 5 moves:

200 runs of 10,000 playouts from opening at 0.5 komi

  mv: C5   count:     3   percent:   1.5000
  mv: D4   count:    27   percent:  13.5000
  mv: D5   count:    96   percent:  48.0000
  mv: E5   count:    71   percent:  35.5000
  mv: C4   count:     3   percent:   1.5000


I'm trying it at 1,000 nodes and get these moves in these proportions
sorted in 349 runs:

 mv: D5   count:    97   percent:  27.7937 
 mv: D4   count:    80   percent:  22.9226
 mv: C4   count:    53   percent:  15.1862
 mv: C5   count:    48   percent:  13.7536
 mv: E5   count:    38   percent:  10.8883
 mv: B5   count:    10   percent:   2.8653
 mv: B4   count:     9   percent:   2.5788
 mv: C3   count:     8   percent:   2.2923
 mv: B3   count:     5   percent:   1.4327
 mv: B2   count:     1   percent:   0.2865

E5 is not even considered best (probably because there are 4 chances to
get D5 for every 1 chance to get E5) but with 100,000 E5 is chosen
almost 90% of the time.

- Don








On Thu, 2008-10-16 at 08:47 -0400, Michael Williams wrote:
> What if you use a faster, lower quality RNG?  How much do the numbers change?
> 
> Don Dailey wrote:
> > Update:
> > 
> > 4173 runs of 100,000 playouts from opening at 0.5 komi
> > 
> > mv: D4   count:     3   percent:   0.0719
> > mv: D5   count:   447   percent:  10.7117
> > mv: E5   count:  3723   percent:  89.2164
> > 
> > 
> >  0.959 percent fall outside the following range ...
> > 
> > score:  lo, med, hi ->       0.52031      0.52433      0.52835
> > nodes:  lo, med, hi ->    11092602.0   11105554.0   11119436.0
> > 
> > 
> > 
> > 
> > On Thu, 2008-10-16 at 00:14 -0400, Don Dailey wrote:
> >> I have some interesting statistics on the simple go program at 0.5 komi
> >> from the starting position.
> >>
> >> I'm running numerous 100,000 game samples and tracking the statistics to
> >> see what kinds of variation I get in scores and nodes. 
> >>
> >> After 1748 runs I see that less than 1 percent of the games score lower
> >> than 0.52027 or higher than 0.52823 when doing 100,000 game playouts.
> >>
> >> So if you get scores outside this range you probably do not have a
> >> conforming program as this is expected to happen less than 1% of the
> >> time.
> >>
> >> But what I really found interested is that only 3 moves (when accounting
> >> for transformations) were chosen from the opening position.  E5 was
> >> chosen 85% of the time,  and most of the remaining time D5 or
> >> equivalent.   Only 1 time was some other move chosen other than these
> >> two and it was D4.
> >>
> >> I wonder how long before it would chose A1?  Probably a very long time
> >> indeed!
> >>
> >> So if your bot chooses a move other than E5 or D5, there is a very good
> >> chance it is not conforming to our specification of a generic MC player.
> >>
> >> ---
> >>
> >>
> >> 1748 runs
> >>
> >> mv: D4   count:     1   percent:   0.0572
> >> mv: D5   count:   200   percent:  11.4416
> >> mv: E5   count:  1547   percent:  88.5011
> >>
> >>
> >>  0.915 percent fall outside the following range ...
> >>
> >> score:  lo, med, hi ->       0.52027      0.52434      0.52823
> >> nodes:  lo, med, hi ->    11093084.0   11105628.5   11119815.0
> >>
> >>
> >> _______________________________________________
> >> computer-go mailing list
> >> [email protected]
> >> http://www.computer-go.org/mailman/listinfo/computer-go/
> >>
> >> ------------------------------------------------------------------------
> >>
> >> _______________________________________________
> >> computer-go mailing list
> >> [email protected]
> >> http://www.computer-go.org/mailman/listinfo/computer-go/
> 
> _______________________________________________
> computer-go mailing list
> [email protected]
> http://www.computer-go.org/mailman/listinfo/computer-go/

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/

Reply via email to