SEQUENTIAL DESIGN OF EXPERIMENTS FOR ESTIMATING QUANTILES OF BLACK-BOX FUNCTIONS

被引：4

作者：

Labopin-Richard, T. ^{[1
]}

Picheny, V. ^{[2
]}

机构：

[1] Univ Paul Sabatier, CNRS UMR 5219, Inst Math Toulouse, 118 Route Narbonne, F-31062 Toulouse, France

[2] Univ Toulouse, MIAT, INRA, Castanet Tolosan, France

来源：

STATISTICA SINICA | 2018年 / 28卷 / 02期

关键词：

Gaussian processes; risk assessment; stepwise uncertainty reduction; COMPUTER EXPERIMENTS; GLOBAL OPTIMIZATION; UNCERTAINTY;

D O I：

10.5705/ss.202016.0160

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Estimating quantiles of black-box deterministic functions with random inputs is a challenging task when the number of function evaluations is severely restricted, which is typical for computer experiments. This article proposes two new sequential Bayesian methods for quantile estimation based on the Gaussian process metamodel. Both rely on the Stepwise Uncertainty Reduction paradigm, hence aim at providing a sequence of function evaluations that reduces an uncertainty measure associated with the quantile estimator. The proposed strategies are tested on several numerical examples, showing that accurate estimators can be obtained using only a small number of function evaluations.

引用

页码：853 / 877

页数：25