[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

21$-66B-51P-42P-42 – Unlimited Games – 3-ply XG2 Rollout

Posted By: Ian Dunstan
Date: Tuesday, 29 July 2014, at 7:51 a.m.

In Response To: 21$-66B-51P-42P-42 – Unlimited Games – 3-ply XG2 Rollout (Timothy Chow)

I'm using "Version GNU Backgammon 1.02.000-mingw 20130728" of gnubg.

I am experimenting with how strong gnubg is vs XG with regard to backgames and containment positions. As an aside to this exercise I noticed how quickly the same (or close to same) standard deviation was reached for a certain position using both bots. It's surprising, but XG seems to take about three times the number of trials to reach the same level of uncertainty gnubg achieves. i.e. that is basically saying that a 3888 trial rollout for XG is the same as a 1296 one for gnubg in terms of reported accuracy.

I have a long rollout nearly finished of a backgame position to support this claim. I will post it in due course.

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.