Asymptotic optimality of myopic information-based strategies for Bayesian adaptive estimation

DOI10.3150/14-BEJ670zbMATH1388.62239WikidataQ110236375MaRDI QIDQ5963513FDOQ5963513

Authors Janne V. Kujala

Publication date 22 February 2016

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1506.05483

zbMATH Keywords

D-optimalitydecision theorysequential estimationasymptotic optimalityactive learningdifferential entropyBayesian adaptive estimationcost of observationactive data selection

Mathematics Subject Classification ID

Bayesian inference (62F15) Applications of statistics to psychology (62P15) Sequential estimation (62L12) Psychophysics and psychophysiology; perception (91E30)

Abstract: This paper presents a general asymptotic theory of sequential Bayesian estimation giving results for the strongest, almost sure convergence. We show that under certain smoothness conditions on the probability model, the greedy information gain maximization algorithm for adaptive Bayesian estimation is asymptotically optimal in the sense that the determinant of the posterior covariance in a certain neighborhood of the true parameter value is asymptotically minimal. Using this result, we also obtain an asymptotic expression for the posterior entropy based on a novel definition of almost sure convergence on "most trials" (meaning that the convergence holds on a fraction of trials that converges to one). Then, we extend the results to a recently published framework, which generalizes the usual adaptive estimation setting by allowing different trial placements to be associated with different, random costs of observation. For this setting, the author has proposed the heuristic of maximizing the expected information gain divided by the expected cost of that placement. In this paper, we show that this myopic strategy satisfies an analogous asymptotic optimality result when the convergence of the posterior distribution is considered as a function of the total cost (as opposed to the number of observations).

Recommendations

• Bayesian adaptive estimation: a theoretical review
• Obtaining the best value for money in adaptive sequential estimation
• Asymptotically pointwise optimal allocation rules in Bayes sequential estimation
• scientific article; zbMATH DE number 1347894
• scientific article; zbMATH DE number 3945174

Cites work

• scientific article; zbMATH DE number 1232374 (Why is no real title available?)
• Asymptotic Statistics
• Asymptotic Theory of Information-Theoretic Experimental Design
• Bayesian adaptive estimation: a theoretical review
• Bayesian adaptive estimation: the next dimension
• Elements of Information Theory
• Obtaining the best value for money in adaptive sequential estimation
• On a Measure of the Information Provided by an Experiment
• On a Strong Law of Large Numbers for Martingales
• Probability Inequalities for Sums of Bounded Random Variables
• Theory of statistics
• Weighted sums of certain dependent random variables

Cited in

(2)

• Bayesian adaptive estimation: a theoretical review
• Information theory for ranking and selection

This page was built for publication: Asymptotic optimality of myopic information-based strategies for Bayesian adaptive estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5963513)