When ignorance is bliss
Peter Grünwald and Joe Halpern
In: UAI 2004, 9-14 July 2004, Banff, Canada.

Abstract

It is commonly-accepted wisdom that more information is better, and that information should never be ignored. Here we argue, using both a Bayesian and a non-Bayesian analysis, that in some situations you are better off ignoring information if your uncertainty is represented by a set of probability measures. These include situations in which the information {\em is\/} relevant for the prediction task at hand. In the non-Bayesian dilation'', the analysis, we show how ignoring information avoids {\em dilation}, the phenomenon that additional pieces of information sometimes lead to an increase in uncertainty. In the Bayesian analysis, we show that for small sample sizes and certain prediction tasks, the Bayesian posterior based on a noninformative prior yields worse predictions than simply ignoring the given information.

 Postscript - Requires a viewer, such as GhostView
EPrint Type: Conference or Workshop Item (Paper) Project Keyword UNSPECIFIED Theory & Algorithms 120 Peter Grünwald 27 May 2004