Comments on: MaxEnt contradicts Bayes Rule? http://hunch.net/?p=209 Machine learning and learning theory research Tue, 07 Sep 2010 23:46:22 -0500 http://wordpress.org/?v=2.8.4 hourly 1 By: David Corfield http://hunch.net/?p=209&cpage=1#comment-24639 David Corfield Sun, 09 Jul 2006 10:58:03 +0000 http://hunch.net/?p=209#comment-24639 That's a coincidence. I just discussed this example on my <a href="http://www.dcorfield.pwp.blueyonder.co.uk/2006/07/conditionalization-as-i-projection.html" rel="nofollow">blog</a>. The key is to see Bayesian updating as just one case of I-projection in the sense of Csiszar. I give a link there to a paper by Harremoes which explains it all. That’s a coincidence. I just discussed this example on my blog. The key is to see Bayesian updating as just one case of I-projection in the sense of Csiszar. I give a link there to a paper by Harremoes which explains it all.

]]> By: Aleks Jakulin http://hunch.net/?p=209&cpage=1#comment-24626 Aleks Jakulin Sun, 09 Jul 2006 04:23:35 +0000 http://hunch.net/?p=209#comment-24626 As with all interpretations and unifications, one just has to pick something that makes most sense. Anyway, a constraint for me is just a parameter in a suitably defined probability distribution. Essentially, E[X] is the parameter, and the distribution is defined to be as the MaxEnt distribution given the value of that parameter. As a Bayesian, however, one would use the data to model the posterior distribution over the parameter. In that sense, I agree with Drew. BTW, another related discussion is at http://groups.google.com/group/sci.stat.math/browse_frm/thread/cb40a8218cb98dfd/4be36e682022d87b As with all interpretations and unifications, one just has to pick something that makes most sense.

Anyway, a constraint for me is just a parameter in a suitably defined probability distribution. Essentially, E[X] is the parameter, and the distribution is defined to be as the MaxEnt distribution given the value of that parameter. As a Bayesian, however, one would use the data to model the posterior distribution over the parameter. In that sense, I agree with Drew.

BTW, another related discussion is at http://groups.google.com/group/sci.stat.math/browse_frm/thread/cb40a8218cb98dfd/4be36e682022d87b

]]>