Efficient Bayesian shape-restricted function estimation with constrained Gaussian process priors

Efficient Bayesian shape-restricted function estimation with constrained Gaussian process priors
复制标题

具有约束高斯过程先验的高效贝叶斯形状限制函数估计

DOI:
10.1007/s11222-020-09922-0
复制
发表时间:
2020
影响因子:
2.2
通讯作者:
Bhattacharya, Anirban
Bhattacharya, Anirban
中科院分区:
数学2区
文献类型:
--
作者:
Ray, Pallavi;Pati, Debdeep;Bhattacharya, Anirban

文献摘要

参考文献

被引文献

相似文献

本文重新审视贝叶斯形状限制推理的问题,在最近开发的近似高斯过程,承认一个等价的配方的形状约束的基础系数。我们提出了一种策略,有效地从所产生的约束后采样吸收asmooth松弛的约束的可能性,并使用循环嵌入技术,从unconstrainedmodified前采样。我们还特别注意减轻高斯过程的协方差核内更新超参数所产生的计算复杂性。在模拟和真实的数据实例中,所开发的算法被证明是准确和高效的。
This article revisits the problem of Bayesian shape-restricted inference in the light of a recently developed approximate Gaussian process that admits an equivalent formulation of the shape constraints in terms of the basis coefficients. We propose a strategy to efficiently sample from the resulting constrained posterior by absorbing asmooth relaxationof the constraint in the likelihood and using circulant embedding techniques to sample from the unconstrainedmodified prior. We additionally pay careful attention to mitigate the computational complexity arising from updating hyperparameters within the covariance kernel of the Gaussian process. The developed algorithm is shown to be accurate and highly efficient in simulated and real data examples.
DOI: 10.1198/073500107000000223
发表时间: 2008-01-01
影响因子: 3
作者:
Brezger, Andreas;Steiner, Winfried J.
通讯作者: Steiner, Winfried J.
DOI: 10.1111/rssb.12162
发表时间: 2017-01-01
影响因子: 5.8
作者:
Botev, Z. I.
通讯作者: Botev, Z. I.
DOI: 10.1007/s11004-017-9673-2
发表时间: 2016-05
影响因子: 2.6
作者:
H. Maatouk;X. Bay
通讯作者: H. Maatouk;X. Bay
DOI: 10.1063/1.3647361
发表时间: 2011
期刊: Zeitschrift für Physik A Atoms and Nuclei
影响因子: --
作者:
J. Bernauer
通讯作者: J. Bernauer
DOI: 10.4103/jpi.jpi_26_20
发表时间: 2021
影响因子: --
作者:
Nearchou IP;Soutar DA;Ueno H;Harrison DJ;Arandjelovic O;Caie PD
通讯作者: Caie PD