Comments on: Probability and Bayesian Epistemology http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/ A general distrust of strong metaphysical claims in mathematics and philosophy. Tue, 14 Jul 2009 21:45:18 +0000 http://wordpress.com/ hourly 1 By: Paul Shearer http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2655 Paul Shearer Fri, 11 Jul 2008 06:48:37 +0000 http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2655 Fun fact from a practicing mathematician -- sometimes in Bayesian inference, "degrees of rational confidence" which are not probability distributions are sometimes used. They are called improper priors. Sometimes improper priors produce paradoxical results though, so they can be modified to produce probability distributions. Fun fact from a practicing mathematician — sometimes in Bayesian inference, “degrees of rational confidence” which are not probability distributions are sometimes used. They are called improper priors.

Sometimes improper priors produce paradoxical results though, so they can be modified to produce probability distributions.

]]> By: Kenny http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2615 Kenny Mon, 24 Mar 2008 16:39:57 +0000 http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2615 I have looked somewhat into fuzzy logic, but I don't think it's really that relevant here. It's more a device for dealing with vagueness than with any of the things that probability is normally taken to deal with. And for the argument that degrees of rational confidence should be modeled with a probability function (or something very much like one), look at <a HREF="http://plato.stanford.edu/entries/epistemology-bayesian/supplement2.html" rel="nofollow">Dutch book arguments</a>. I have looked somewhat into fuzzy logic, but I don’t think it’s really that relevant here. It’s more a device for dealing with vagueness than with any of the things that probability is normally taken to deal with. And for the argument that degrees of rational confidence should be modeled with a probability function (or something very much like one), look at Dutch book arguments.

]]> By: Charles Wells http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2614 Charles Wells Fri, 21 Mar 2008 20:18:41 +0000 http://antimeta.wordpress.com/2007/12/10/probability-and-bayesian-epistemology/#comment-2614 Taking degrees of rational confidence as given sounds more like fuzzy logic than probability. Have you looked into that? Taking degrees of rational confidence as given sounds more like fuzzy logic than probability. Have you looked into that?

]]>