<div dir="ltr"><div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small;color:rgb(0,0,0)">noninformative priors: is that like committing crimes or marriages without learning anything from from them?<br><br></div><div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small;color:rgb(0,0,0)">bill w<br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Sun, Nov 22, 2015 at 1:02 AM, Anders Sandberg <span dir="ltr"><<a href="mailto:anders@aleph.se" target="_blank">anders@aleph.se</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000"><span class="">
On 2015-11-22 04:44, spike wrote:<br>
<blockquote type="cite">
<div><br>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1f497d">It
occurred to me that without some kind of indication of a
range of possible amounts in the vault, this is any ordinary
trivial guessing game. But if we assume the amount of money
in the vault is somewhere between 0 and 6,283,185 dollars
with each amount being equally probable, the problem becomes
a calculation.</span></p>
</div>
</blockquote>
<br></span>
I actually liked the undefined nature of the amount. Thinking about
things that could have any order of magnitude brings up really cool
issues of noninformative priors.<br>
<br>
The thing reminds me of the problem of doing a Bayesian analysis of
the German tank problem when you have only one tank (it came up
during I lecture I gave last week). <br>
<a href="https://en.wikipedia.org/wiki/German_tank_problem" target="_blank">https://en.wikipedia.org/wiki/German_tank_problem</a><br>
Using a noninformative prior there will not work for one tank. If
you have a known upper limit Omega on how many tanks there could be,
then you can assume the actual number is uniform between 1 and Omega
and do a calculation, ending up with an estimate.<br>
<br>
In your lovely problem having an upper limit Omega makes things
easier but less interesting. <br><span class="HOEnZb"><font color="#888888">
<pre cols="72">--
Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University</pre>
</font></span></div>
<br>_______________________________________________<br>
extropy-chat mailing list<br>
<a href="mailto:extropy-chat@lists.extropy.org">extropy-chat@lists.extropy.org</a><br>
<a href="http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat" rel="noreferrer" target="_blank">http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat</a><br>
<br></blockquote></div><br></div>