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

Download:

Current browse context:

stat.CO

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

Statistics > Methodology

Title: Efficient methods for Gaussian Markov random fields under sparse linear constraints

Abstract: Methods for inference and simulation of linearly constrained Gaussian Markov Random Fields (GMRF) are computationally prohibitive when the number of constraints is large. In some cases, such as for intrinsic GMRFs, they may even be unfeasible. We propose a new class of methods to overcome these challenges in the common case of sparse constraints, where one has a large number of constraints and each only involves a few elements. Our methods rely on a basis transformation into blocks of constrained versus non-constrained subspaces, and we show that the methods greatly outperform existing alternatives in terms of computational cost. By combining the proposed methods with the stochastic partial differential equation approach for Gaussian random fields, we also show how to formulate Gaussian process regression with linear constraints in a GMRF setting to reduce computational cost. This is illustrated in two applications with simulated data.
Subjects: Methodology (stat.ME); Computation (stat.CO); Machine Learning (stat.ML)
MSC classes: 62M40, 60G15, 65C30
ACM classes: G.3; G.1.3
Cite as: arXiv:2106.01712 [stat.ME]
  (or arXiv:2106.01712v1 [stat.ME] for this version)

Submission history

From: Jonas Wallin [view email]
[v1] Thu, 3 Jun 2021 09:31:12 GMT (54kb,D)

Link back to: arXiv, form interface, contact.