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

BGonline.org Forums

Using two bots instead of a longer rollout: effect on confidence in result

Posted By: leobueno
Date: Saturday, 22 December 2012, at 6:43 p.m.

In Response To: Using two bots instead of a longer rollout: effect on confidence in result (Tom Keith)

Tom Keith said:

An 89.9% chance of winning a rollout is the same as a JSD of about 1.28. (To get this number I consulted the table here: http://www.bkgm.com/rgb/rgb.cgi?view+1238)

You can combine two independent JSDs using the formula below. Here JSD1 = 1.28 (from above) and JSD2 = 1.803 (from your GNU rollout).

CombinedJSD = sqrt( JSD1^2 + JSD2^2 )

= sqrt( 1.28^2 + 1.803^2 )

= sqrt( 1.638 + 3.251 )

= sqrt( 4.889 )

= 2.21

Leo Bueno asks:

So, if I stop XG's rollouts when the top move has 99.0% chance to be the best--which corresponds in the table to a 2.4 JSD--and I stop the GNUbg rollouts when JSDs from best choice is also 2.4, then the combined JSD (cJSD) will be

cJSD = sqrt (2.4^2 + 2.4^2)

= sqrt (5.76 + 5.76)

= sqrt (11.52)

= 3.39

That would give me a fairly high cJSD.

I also like the idea of having had "two opinions" from two separate bots, so that when they I agree, I'll have a warmer and fuzzier feeling about the best play, much like when I get a second medical opinion before deciding on a course of treatment.

I am wondering why Bob Koca thinks this is not a reasonable approach to rolling out plays, preferring longer XG rollouts.

Also, does a combined JSD of 3.39 overcome Tim Chow's concern about stopping rollouts at 3 JSDs (http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=63159)?

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.