For my fast/dumb neural net engine, Antbot9x9, I coevolved the weights using a 
similar tournament system. Each individual played a number of games against all 
the others, round robin, and the score was the sum of points for all of its 
games.

Some observations/claims:
Non-transitive effects seem more visible. Consistently overplaying garners 
extra points from weak opponents but needlessly loses extra points against 
strong ones. It becomes more important to play your opponent as well as the 
board: if you think that you have him outmatched, take some risky gambles, 
overplay. Every game in the tournament matters, right till the end of that game.

I think it could be interesting to try some bot tournaments like this. It might 
be fun to watch. When the strongest bot was playing the weakest, even near the 
(painfully one-sided) end of the game there would be an element of suspense. 
The stronger bot would (or should) be trying to swindle a few last extra points 
it didn't deserve, and the fate of the tournament could hinge on it.

- Dave Hillis


-----Original Message-----
From: "Ingo Althöfer" <3-hirn-ver...@gmx.de>
To: computer-go@computer-go.org
Sent: Mon, Nov 23, 2009 4:10 am
Subject: [computer-go] Re: Hahn system tournament and MC bots


Alain Baeckeroot wrote:
 A Go tounrmaent with Hahn system has been retransmeted 
 see  ... http://www.suomigo.net/wiki/HahnSystem
Thanks for the interesting stuff and the links.
>From the link HahnSystem:
 Winning By 0.5-10       gets  60 points
 Winning by 10.5-20      gets  70 points
 Winning by 20.5-30      gets  80 points
 Winning by 30.5-40      gets  90 points
 Winning by >40.5        gets 100 points

 Losing by  0.5-10       gets  40 points
 Losing by  10.5-20      gets  30 points
 Losing by  20.5-30      gets  20 points
 Losing by  30.5-40      gets  10 points
 Losing by  >40.5        gets   0 point
I would have found a "completely continuous result system"
ore natural, for instance
iving +40.5 points for each win with 40.5 or more
iving -40.5 points for each loss with 40.5 or more
iving x points for each score in between +40.5 and -40.5
And I believe that MC-bots would gain about 2 rating levels
gainst human players, when compared with normal chinese scoring.
Just my 40.5 cent, Ingo.
-- 
reisknaller: GMX DSL Flatrate für nur 16,99 Euro/mtl.!
ttp://portal.gmx.net/de/go/dsl02
______________________________________________
omputer-go mailing list
omputer...@computer-go.org
ttp://www.computer-go.org/mailman/listinfo/computer-go/

_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/

Reply via email to