arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.19977stat.ME
收藏

基于多分辨率预测过程的可扩展和鲁棒空间预测

Scalable and Robust Spatial Prediction via Multi-Resolution Ensembles of Predictive Processes

Scientific Computing Center, Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院科学计算中心)

Nicolas Bianco, Nadja Klein

AI总结:

本文提出基于空间划分的预测过程集合,通过多分辨率方法实现空间预测的可扩展性和鲁棒性,提升计算效率与预测精度。

摘要:

高斯过程提供了一种灵活的空间预测框架,但其计算成本限制了大规模数据的应用。预测过程(PPs)通过将原始过程投影到少量诱导点上,降低了计算负担。然而,现有理论要求诱导点数量随样本量增长,导致精度与效率之间的权衡。本文通过引入基于空间划分的PPs集合,提出新的划分与拼接方案。通过推广PPs的收敛结果,能够显式平衡可扩展性与精度:增加集合组件数量会减缓收敛速度,但显著提高计算效率。进一步理论证明,尽管PPs在固定诱导点数量下存在有限的近似精度,但它们在数据污染下是渐近稳健的。受此启发,本文最终引入多分辨率集合,结合固定诱导点数量的PPs与多个可能重叠的粗到细划分定义的集合。模拟和大规模地统计应用表明,本文方法在准确性和鲁棒性方面表现优异,同时提供计算优势,为空间预测提供实用且广泛适用的解决方案。

英文摘要:

Gaussian processes provide a flexible framework for spatial prediction, but their computational cost limits applicability to large-scale data with large sample size $n$. Predictive processes (PPs), a popular low-rank approximation, mitigate this burden by projecting the original process onto a reduced set of $m\ll n$ inducing points. However, existing theory requires $m$ to grow with $n$, creating a trade-off between accuracy and computational efficiency. We address this challenge by introducing an ensemble of PPs based on spatial partitioning, and propose a novel partitioning and patching scheme with desirable properties. By generalizing the convergence results of PPs, it becomes possible to explicitly balance scalability and accuracy: increasing the number of ensemble components slows down the convergence but substantially improves computational efficiency. We further show theoretically that, despite the limited approximation accuracy of PPs with fixed, small $m$, they provide predictions that are more robust to training data contamination. Motivated by these findings, we finally introduce a multi-resolution ensemble that combines ensembles over possibly overlapping coarse to fine partitions. Unlike previous multi-resolution approaches, our method targets predictive robustness rather than accurate covariance approximation. Simulations and large-scale geostatistical applications demonstrate that our approach delivers accurate, robust predictions while being computationally efficient, thereby providing a practical and broadly applicable solution for spatial prediction.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑