[Top][All Lists]

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

Re: [Bug-gnubg] Suggestion: equivalent # of games played in roll outs th

From: Øystein O Johansen
Subject: Re: [Bug-gnubg] Suggestion: equivalent # of games played in roll outs thx to var. reduction
Date: Tue, 14 Oct 2003 16:28:37 +0200

> I'm aware that the issue was raised in the past various
> times, I'd just like to know the current status, if any :

> - is it on the agenda at all?

I tried to implement this some (years?) ago, but I found out that the
standard error tells you much more than the equivalent # of games, so I
rejected the code.

> - is it on the agenda but with a very low priority level?
> (which would perfectly make sense, regarding a more or
> less cosmetic feature)

No, I don't think it's on the agenda.

> - is it hard to compute/implement?

No, actually it's very easy. Here's how you can compute it on your pocket

                        var(wo. vr)
Equivalent # of games = ------------ x real number of games
                        var(w. vr)


var(w. vr) = the variance with variance reduction = sdt.err^2
var(wo vr) = the variance without variance reduction.
             (Calculated based on the results)

> Incidentally, on the 'promotion' side (if ever needed)
> I think this would show even better the top quality of
> roll outs enabled by Gnubg

Maybe, but I don't worry... ;-)


The information contained in this message may be CONFIDENTIAL and is
intended for the addressee only. Any unauthorised use, dissemination of the
information or copying of this message is prohibited. If you are not the
addressee, please notify the sender immediately by return e-mail and delete
this message.
Thank you.

reply via email to

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