We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.ST

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Statistics Theory

Title: Radial Neighbors for Provably Accurate Scalable Approximations of Gaussian Processes

Abstract: In geostatistical problems with massive sample size, Gaussian processes (GP) can be approximated using sparse directed acyclic graphs to achieve scalable $O(n)$ computational complexity. In these models, data at each location are typically assumed conditionally dependent on a small set of parents which usually include a subset of the nearest neighbors. These methodologies often exhibit excellent empirical performance, but the lack of theoretical validation leads to unclear guidance in specifying the underlying graphical model and may result in sensitivity to graph choice. We address these issues by introducing radial neighbors Gaussian processes and corresponding theoretical guarantees. We propose to approximate GPs using a sparse directed acyclic graph in which a directed edge connects every location to all of its neighbors within a predetermined radius. Using our novel construction, we show that one can accurately approximate a Gaussian process in Wasserstein-2 distance, with an error rate determined by the approximation radius, the spatial covariance function, and the spatial dispersion of samples. Our method is also insensitive to specific graphical model choice. We offer further empirical validation of our approach via applications on simulated and real world data showing state-of-the-art performance in posterior inference of spatial random effects.
Subjects: Statistics Theory (math.ST); Methodology (stat.ME)
Cite as: arXiv:2211.14692 [math.ST]
  (or arXiv:2211.14692v3 [math.ST] for this version)

Submission history

From: Yichen Zhu [view email]
[v1] Sun, 27 Nov 2022 00:57:29 GMT (1580kb,D)
[v2] Tue, 6 Dec 2022 01:56:22 GMT (1579kb,D)
[v3] Thu, 8 Dec 2022 04:01:51 GMT (1579kb,D)

Link back to: arXiv, form interface, contact.