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

BGonline.org Forums

Rollouts - Long

Posted By: Rich Munitz
Date: Thursday, 1 January 2009, at 10:49 p.m.

In Response To: Rollouts - Long (David Rockwell)

I would also be just as agreeable that the different seeds are not the cause of the discrepancy. Yet, the 95% confidence interval suggests that the differences are not caused by chance alone.

I don't use GNU, but I am curious. If we look at the plays made with each setting for the 21 dice rolls in reply to the two opening 64 plays, does GNU play the same in all cases? It seems likely that any major variation in absolute equity is likely to be from play differences in the first ply, whereas play differences in later plies should be more subtle.

I am also not an expert on variance reduction, but I've always wondered whether the "adjustments" introduce bias into the results due to misunderstandings of the equities. In other words, does it tend to slightly ignore the actual results and push the rollout results towards its "expected" result. Perhaps in a short trial, the variance reduction gives a better result as it reduces the dominant effect of the small sample size. But maybe with a large sample size, this bias may start to dominate. Sorry if I am just being ignorant here. Can't know everything.

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.