A Locally Adaptive Bayesian Cubature Method

A Locally Adaptive Bayesian Cubature Method
复制标题

局部自适应贝叶斯体积法

DOI:
--
复制
发表时间:
2019
期刊:
International Conference on Artificial Intelligence and Statistics
影响因子:
--
通讯作者:
A. Teckentrup
A. Teckentrup
中科院分区:
--
文献类型:
--
作者:
Matthew A. Fisher;C. Oates;C. Powell;A. Teckentrup

文献摘要

参考文献

被引文献

相似文献

贝叶斯体积 (BC) 是一种关于昂贵被积函数体积的流行推理视角,其中被积函数使用随机过程模型进行模拟。已经提出了几种方法将顺序适应(即依赖于先前的被积函数评估)编码到该框架中。然而,这些建议仅限于估计平稳协方差模型的参数或将计算资源集中在被积函数取大值的区域。相比之下,许多经典的自适应体积方法将计算资源集中在局部误差估计最大的空间区域。这项工作的贡献有三方面:首先,我们提出了一个理论结果,表明不存在经典自适应梯形方法的直接贝叶斯类似物。然后我们提出了一种新颖的 BC 方法,该方法在经验上与自适应梯形方法具有相似的行为。最后,我们通过详细的实证评估证明,相对于标准 BC,该新方法提供了改进的体积性能。
Bayesian cubature (BC) is a popular inferential perspective on the cubature of expensive integrands, wherein the integrand is emulated using a stochastic process model. Several approaches have been put forward to encode sequential adaptation (i.e. dependence on previous integrand evaluations) into this framework. However, these proposals have been limited to either estimating the parameters of a stationary covariance model or focusing computational resources on regions where large values are taken by the integrand. In contrast, many classical adaptive cubature methods focus computational resources on spatial regions in which local error estimates are largest. The contributions of this work are three-fold: First, we present a theoretical result that suggests there does not exist a direct Bayesian analogue of the classical adaptive trapezoidal method. Then we put forward a novel BC method that has empirically similar behaviour to the adaptive trapezoidal method. Finally we present evidence that the novel method provides improved cubature performance, relative to standard BC, in a detailed empirical assessment.
DOI: --
发表时间: 2017-11
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Matthew M. Dunlop;M. Girolami;A. Stuart;A. Teckentrup
通讯作者: Matthew M. Dunlop;M. Girolami;A. Stuart;A. Teckentrup
概率数值学和计算中的不确定性。
DOI: 10.1098/rspa.2015.0142
发表时间: 2015-07-08
期刊: Proceedings. Mathematical, physical, and engineering sciences
影响因子: --
作者:
Hennig P;Osborne MA;Girolami M
通讯作者: Girolami M
用于高效、非近视顺序实验设计的双筒望远镜
DOI: --
发表时间: 2020
期刊: Proceedings of the 37th International Conference on Machine Learning
影响因子: --
作者:
Jiang, Shali;Chai, Henry;González, Javier;Garnett, Roman
通讯作者: Garnett, Roman