Distributed nearest-neighbor Gaussian processes

Distributed nearest-neighbor Gaussian processes
复制标题

分布式最近邻高斯过程

DOI:
10.1080/03610918.2021.1921798
复制
发表时间:
2021
期刊:
Communications in Statistics - Simulation and Computation
影响因子:
--
通讯作者:
Sansó, Bruno
Sansó, Bruno
中科院分区:
--
文献类型:
--
作者:
Grenier, Isabelle;Sansó, Bruno

文献摘要

参考文献

相似文献

虽然许多统计方法已经解决了大型空间数据集的问题,但由于昂贵的数据移动和数据存储而产生的问题长期被搁置在一边。易于访问数据已被认为是理所当然的,现在正成为统计推断性能的一个重要瓶颈。随着高分辨率空间数据的可用性不断增长,开发利用多处理器和多存储功能的高效建模技术的需求正在成为一个优先事项。为此,开发一种分布式方法来实现最近邻高斯过程(NNGP)模型,用于大数据集的空间插值和推理。提出的框架保留了NNGP的精确实现,同时允许后验推理的分布式或顺序计算。该方法允许对数据进行任意分组,无论是随机分组还是按区域分组。由于这种新方法,NNGP模型可以在主节点级别以最小的过载实现计算负担的平均分配。
While many statistical approaches have tackled the problem of large spatial datasets, the issues arising from costly data movement and data storage have long been set aside. Having easy access to the data has been taken for granted and is now becoming an important bottleneck in the performance of statistical inference. As the availability of high resolution spatial data continues to grow, the need to develop efficient modeling techniques that leverage multi-processor and multi-storage capabilities is becoming a priority. To that end, the development of a distributed method to implement Nearest-Neighbor Gaussian Process (NNGP) models for spatial interpolation and inference for large datasets is of interest. The proposed framework retains the exact implementation of the NNGP while allowing for distributed or sequential computation of the posterior inference. The method allows for any choice of grouping of the data whether it is at random or by region. As a result of this new method, the NNGP model can be implemented with an even split of the computation burden with minimum overload at the master node level.
DOI: 10.1080/00401706.2015.1102763
发表时间: 2017-01
期刊: Technometrics
影响因子: 2.5
作者:
Matthew J. Heaton;W. F. Christensen;Maria A. Terres
通讯作者: Matthew J. Heaton;W. F. Christensen;Maria A. Terres
DOI: 10.1214/17-ba1083
发表时间: 2017-11
期刊: Bayesian Analysis
影响因子: 4.4
作者:
Hang Qian
通讯作者: Hang Qian
DOI: 10.1214/17-ba1056r
发表时间: 2017-06
期刊: Bayesian analysis
影响因子: 4.4
作者:
Banerjee S
通讯作者: Banerjee S
DOI: 10.1007/s13253-018-00348-w
发表时间: 2019-09-01
影响因子: 1.4
作者:
Heaton, Matthew J.;Datta, Abhirup;Zammit-Mangion, Andrew
通讯作者: Zammit-Mangion, Andrew
DOI: 10.1080/10618600.2018.1537924
发表时间: 2019-03-21
影响因子: 2.4
作者:
Finley, Andrew O.;Datta, Abhirup;Banerjee, Sudipto
通讯作者: Banerjee, Sudipto