Bayesian quadrature optimization (BQO) maximizes the expectation of an
expensive black-box integrand taken over a known probability distribution. In
this work, we study BQO under distributional uncertainty in which the
underlying probability distribution is unknown except for a limited set of its
i.i.d. samples. A standard BQO approach maximizes the Monte Carlo estimate of
the true expected objective given the fixed sample set. Though Monte Carlo
estimate is unbiased, it has high variance given a small set of samples; thus
can result in a spurious objective function. We adopt the distributionally
robust optimization perspective to this problem by maximizing the expected
objective under the most adversarial distribution. In particular, we propose a
novel posterior sampling based algorithm, namely distributionally robust BQO
(DRBQO) for this purpose. We demonstrate the empirical effectiveness of our
proposed framework in synthetic and real-world problems, and characterize its
theoretical convergence via Bayesian regret.
%0 Journal Article
%1 nguyen2020distributionally
%A Nguyen, Thanh Tang
%A Gupta, Sunil
%A Ha, Huong
%A Rana, Santu
%A Venkatesh, Svetha
%D 2020
%K bayesian optimization readings robustness
%T Distributionally Robust Bayesian Quadrature Optimization
%U http://arxiv.org/abs/2001.06814
%X Bayesian quadrature optimization (BQO) maximizes the expectation of an
expensive black-box integrand taken over a known probability distribution. In
this work, we study BQO under distributional uncertainty in which the
underlying probability distribution is unknown except for a limited set of its
i.i.d. samples. A standard BQO approach maximizes the Monte Carlo estimate of
the true expected objective given the fixed sample set. Though Monte Carlo
estimate is unbiased, it has high variance given a small set of samples; thus
can result in a spurious objective function. We adopt the distributionally
robust optimization perspective to this problem by maximizing the expected
objective under the most adversarial distribution. In particular, we propose a
novel posterior sampling based algorithm, namely distributionally robust BQO
(DRBQO) for this purpose. We demonstrate the empirical effectiveness of our
proposed framework in synthetic and real-world problems, and characterize its
theoretical convergence via Bayesian regret.
@article{nguyen2020distributionally,
abstract = {Bayesian quadrature optimization (BQO) maximizes the expectation of an
expensive black-box integrand taken over a known probability distribution. In
this work, we study BQO under distributional uncertainty in which the
underlying probability distribution is unknown except for a limited set of its
i.i.d. samples. A standard BQO approach maximizes the Monte Carlo estimate of
the true expected objective given the fixed sample set. Though Monte Carlo
estimate is unbiased, it has high variance given a small set of samples; thus
can result in a spurious objective function. We adopt the distributionally
robust optimization perspective to this problem by maximizing the expected
objective under the most adversarial distribution. In particular, we propose a
novel posterior sampling based algorithm, namely distributionally robust BQO
(DRBQO) for this purpose. We demonstrate the empirical effectiveness of our
proposed framework in synthetic and real-world problems, and characterize its
theoretical convergence via Bayesian regret.},
added-at = {2020-06-09T19:14:14.000+0200},
author = {Nguyen, Thanh Tang and Gupta, Sunil and Ha, Huong and Rana, Santu and Venkatesh, Svetha},
biburl = {https://www.bibsonomy.org/bibtex/23416eebce636e76acbb008d8dbe12f05/kirk86},
description = {[2001.06814] Distributionally Robust Bayesian Quadrature Optimization},
interhash = {2fcb7e7cf27b4d6bc1ae10c97a92838f},
intrahash = {3416eebce636e76acbb008d8dbe12f05},
keywords = {bayesian optimization readings robustness},
note = {cite arxiv:2001.06814Comment: AISTATS2020},
timestamp = {2020-06-09T19:14:44.000+0200},
title = {Distributionally Robust Bayesian Quadrature Optimization},
url = {http://arxiv.org/abs/2001.06814},
year = 2020
}