I post here a reproducible case where 0.15 rollouts perform
significantly poorer than 0.14. I have 2 .sgf files with the rollout
results if you would like me to send them. I say poorer because it is
essentially a one-sided checkerplay position.
If I get time I could also try this with a larger search space for the
pruning net but I wanted to have identical settings for both nets
initially.
I rolled this out with 0.14 2-ply and the wins were 75.1% with gammon
losses at least 4% lower, and the standard error was 0.001 wins, and
0.006 cubeful (one-third the error of the 0.15 version) after only 600+
trials. I was unable to save that 2-ply rollout result, so I am
repeating the 0.14 rollout today and could forward the .sgf if you like.
Here is a discussion thread at GOL about this position with several
additional rollout results including 0-ply GNU and Snowie 4.1.
http://www.gammonline.com/members/board/config.cgi?read=85765
My general observation after several rollouts with 0.15 is that the
error rates for cubeful equities are taking several times more trials to
converge than the 0.14 net.
This is a general observation as an experienced user, not a claim as a
result of any scientific study.
While we can all enjoy the immediate benefits of match analysis that is
twice as fast, it could just be that it will take us some time to
rediscover the ideal rollout settings with the 0.15 version.
--Ned
On Sat, 16 Oct 2004 12:42:24 +1300
Joseph Heled <address@hidden> wrote:
... Please post a reproducible case, with a smaller n as possible,
full setting, and seed ...
-Joseph
------------------------------------------------------------------------
The score (after 0 games) is: White 0, Blue 0 (match to 5 points)
Match Information:
Date: October 22, 2004
Move number 7: X to play 62
GNU Backgammon Position ID: 7wcAAMK2DRDYAA
Match ID : UQmrAAAAAAAA
+13-14-15-16-17-18------19-20-21-22-23-24-+ O: White
| X | | X X O O | O 0 points
| | | X X O O | O
| | | O O | O
| | | O O | O
| | | 6 |
v| |BAR| | 5 point match
| | | |
| | | |
| | | |
| X | | X X X X | Rolled 62
| X | | X X X X O | 0 points
+12-11-10--9--8--7-------6--5--4--3--2--1-+ X: Blue (Cube: 2)
Pip counts: O 40, X 151
* Blue doubles
Alert: wrong double ( -0.021)!
Cube analysis
Rollout cubeless equity +0.331 (Money: +0.309)
Cubeful equities:
1. No double +0.816
2. Double, pass +1.000 ( +0.184)
3. Double, take +0.794 ( -0.021)
Proper cube action: No redouble, take (10.4%)
Rollout details:
Player Blue owns 2-cube:
0.715 0.000 0.000 - 0.285 0.118 0.003 CL +0.331 CF +0.816
[0.003 0.000 0.000 - 0.003 0.003 0.001 CL 0.008 CF 0.014]
Player White owns 4-cube:
0.714 0.000 0.000 - 0.286 0.121 0.006 CL +0.876 CF +0.794
[0.003 0.000 0.000 - 0.003 0.003 0.001 CL 0.017 CF 0.017]
Full cubeful rollout with var.redn.
2592 games, Mersenne Twister dice gen. with seed 945052864 and quasi-random dice
Play: world class 2-ply cubeful prune [world class]
keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
Skip pruning for 1-ply moves.
Cube: 2-ply cubeful prune [world class]
Output generated Tue Oct 26 10:34:05 2004
by GNU Backgammon 0.14-devel (Text Export version 1.68)