CaltechAUTHORS
  A Caltech Library Service

Brittleness of Bayesian inference and new Selberg formulas

Owhadi, Houman and Scovel, Clint (2015) Brittleness of Bayesian inference and new Selberg formulas. . (Submitted) https://resolver.caltech.edu/CaltechAUTHORS:20160224-073833523

[img] PDF - Submitted Version
See Usage Policy.

502Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20160224-073833523

Abstract

The incorporation of priors in the Optimal Uncertainty Quantification (OUQ) framework reveals brittleness in Bayesian inference; a model may share an arbitrarily large number of finite-dimensional marginals with, or be arbitrarily close (in Prokhorov or total variation metrics) to, the data-generating distribution and still make the largest possible prediction error after conditioning on an arbitrarily large number of samples. The initial purpose of this paper is to unwrap this brittleness mechanism by providing (i) a quantitative version of the Brittleness Theorem of and (ii) a detailed and comprehensive analysis of its application to the revealing example of estimating the mean of a random variable on the unit interval [0, 1] using priors that exactly capture the distribution of an arbitrarily large number of Hausdorff moments. However, in doing so, we discovered that the free parameter associated with Markov and Kreĩn’s canonical representations of truncated Hausdorff moments generates reproducing kernel identities corresponding to reproducing kernel Hilbert spaces of polynomials. Furthermore, these reproducing identities lead to biorthogonal systems of Selberg integral formulas. This process of discovery appears to be generic: whereas Karlin and Shapley used Selberg’s integral formula to first compute the volume of the Hausdorff moment space (the polytope defined by the first n moments of a probability measure on the interval [0, 1]), we observe that the computation of that volume along with higher order moments of the uniform measure on the moment space, using different finite-dimensional representations of subsets of the infinite-dimensional set of probability measures on [0, 1] representing the first n moments, leads to families of equalities corresponding to classical and new Selberg identities.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
http://arxiv.org/abs/1304.7046arXivDiscussion Paper
ORCID:
AuthorORCID
Owhadi, Houman0000-0002-5677-1600
Scovel, Clint0000-0001-7757-3411
Additional Information:(Submitted on 26 Apr 2013 (v1), last revised 24 Oct 2014 (this version, v2)). October 27, 2014. We would like to thank Gérard Letac for his helpful comments, in particular for his substantial simplification, included here, of our previous proof of Lemma 4.1. We would also like to thank one of the referees for many helpful comments which we also feel improved the manuscript. The authors gratefully acknowledge this work supported by the Air Force Office of Scientific Research under Award Number FA9550-12-1-0389 (Scientific Computation of Optimal Statistical Estimators).
Funders:
Funding AgencyGrant Number
Air Force Office of Scientific Research (AFOSR)FA9550-12-1-0389
Subject Keywords:Bayesian inference, misspecification, robustness, uncertainty quantification, optimal uncertainty quantification, reproducing kernel Hilbert spaces (RKHS), Selberg integral formulas
Classification Code:2010 Mathematics Subject Classification: 62A01, 62F12, 62F15, 62G20, 62G35. 46E22, 11M36
Record Number:CaltechAUTHORS:20160224-073833523
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20160224-073833523
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:64713
Collection:CaltechAUTHORS
Deposited By: Ruth Sustaita
Deposited On:24 Feb 2016 18:12
Last Modified:03 Oct 2019 09:40

Repository Staff Only: item control page