[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug-gnubg] Best Rollout Parameters?

From: crawling . weasel
Subject: [Bug-gnubg] Best Rollout Parameters?
Date: Fri, 13 Aug 2004 14:31:15 +0200


What are the best (1st: reliable (!!!) 2nd: as fast as possible) parameters for 
rollouts, please? 

Years ago Gary Wong recommended 1296 trials, 2 ply, no truncation and variance 
reduction. Still 
the same? 2ply seems a lot and slow ...

Meanwhile there are a lot more parameters. Should I use:

"Stop when STD are small enough"?
"Stop rollout of multiple moves based on j.s.d."?

Anything else important? 

Thanks a lot in advance.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]