Optimal learning under robustness and time-consistency

DOI10.1287/OPRE.2019.1899zbMATH Open1497.91086arXiv1708.01890OpenAlexW2919407062MaRDI QIDQ5095142FDOQ5095142

Authors: Larry G. Epstein, Shaolin Ji 

Publication date: 5 August 2022

Published in: Operations Research (Search for Journal in Brave)

Abstract: We model learning in a continuous-time Brownian setting where there is prior ambiguity. The associated model of preference values robustness and is time-consistent. It is applied to study optimal learning when the choice between actions can be postponed, at a per-unit-time cost, in order to observe a signal that provides information about an unknown parameter. The corresponding optimal stopping problem is solved in closed-form, with a focus on two specific settings: Ellsberg's two-urn thought experiment expanded to allow learning before the choice of bets, and a robust version of the classical problem of sequential testing of two simple hypotheses about the unknown drift of a Wiener process. In both cases, the link between robustness and the demand for learning is studied.

Full work available at URL: https://arxiv.org/abs/1708.01890

Recommendations

• Robust experimentation in the continuous time bandit problem
• Robustness and ambiguity in continuous time
• Optimal stopping under ambiguity in continuous time
• Learning Under Ambiguity
• Robust best choice problem

zbMATH Keywords

learningmodel uncertaintyambiguityoptimal stoppingpartial informationrecursive utilitytime-consistencyEllsberg paradoxrobust decisionssequential testing of simple hypotheses

Mathematics Subject Classification ID

Decision theory (91B06)

Cites Work

• Sequential Tests of Statistical Hypotheses
• Statistical decision theory and Bayesian analysis. 2nd ed
• Title not available (Why is that?)
• Title not available (Why is that?)
• A Robust Version of the Probability Ratio Test
• Robust Bayesian analysis
• An overview of robust Bayesian analysis. (With discussion)
• Maxmin expected utility with non-unique prior
• Reflected solutions of backward SDE's, and related obstacle problems for PDE's
• Optimal stopping under ambiguity in continuous time
• Risk, ambiguity and the Savage axioms
• Title not available (Why is that?)
• Ambiguity, Risk, and Asset Returns in Continuous Time
• Recursive multiple-priors.
• Title not available (Why is that?)
• Learning Under Ambiguity
• Theory of decision under uncertainty
• Rationality of belief or: why Savage's axioms are neither necessary nor sufficient for rationality
• Bayes and Minimax Solutions of Sequential Decision Problems
• Reasoning to a Foregone Conclusion
• Robust control of the multi-armed bandit problem
• The K-armed bandit problem with multiple priors
• Optimal Stopping Rules
• Learning from ambiguous urns
• Rectangular sets of probability measures
• Backward stochastic differential equations. From linear to fully nonlinear theory
• Shunning uncertainty: the neglect of learning opportunities
• General stopping behaviors of naïve and noncommitted sophisticated agents, with application to probability distortion

Cited In (5)

• Optimal and robust memoryless discrimination from dependent observations
• On ε-Optimality of the Pursuit Learning Algorithm
• Robust experimentation in the continuous time bandit problem
• Robustness and ambiguity in continuous time
• Robust Partial-Learning in Linear Gaussian Systems

This page was built for publication: Optimal learning under robustness and time-consistency

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5095142)