Correlation-based sparse inverse Cholesky factorization for fast Gaussian-process inference
Preprint
Abstract
Gaussian processes are widely used as priors for unknown functions in statistics and machine learning. To achieve computationally feasible inference for large datasets, a popular approach is the Vecchia approximation, which is an ordered conditional approximation of the data vector that implies a sparse Cholesky factor of the precision matrix. The ordering and sparsity pattern are typically determined based on Euclidean distance of the inputs or locations corresponding to the data points. Here, we propose instead to use a correlation-based distance metric, which implicitly applies the Vecchia approximation in a suitable transformed input space. The correlation-based algorithm can be carried out in quasilinear time in the size of the dataset, and so it can be applied even for iterative inference on unknown parameters in the correlation structure. The correlation-based approach has two advantages for complex settings: It can result in more accurate approximations, and it offers a simple, automatic strategy that can be applied to any covariance, even when Euclidean distance is not applicable. We demonstrate these advantages in several settings, including anisotropic, nonstationary, multivariate, and spatio-temporal processes. We also illustrate our method on multivariate spatio-temporal temperature fields produced by a regional climate model.Keywords
All Related Versions
This publication has 41 references indexed in Scilit:
- Gaussian process learning via Fisher scoring of Vecchia’s approximationStatistics and Computing, 2021
- High‐dimensional multivariate geostatistics: A Bayesian matrix‐normal approachEnvironmetrics, 2021
- Vecchia Approximations of Gaussian-Process PredictionsJournal of Agricultural, Biological and Environmental Statistics, 2020
- Computationally efficient nonstationary nearest‐neighbor Gaussian process models using data‐driven techniquesEnvironmetrics, 2019
- Nonseparable covariance models on circles cross time: A study of Mexico City ozoneEnvironmetrics, 2019
- A Case Study Competition Among Methods for Analyzing Large Spatial DataJournal of Agricultural, Biological and Environmental Statistics, 2018
- On the non-reducibility of non-stationary correlation functions to stationary ones under a class of mean-operator transformationsStochastic Environmental Research and Risk Assessment, 2009
- On the Use of Non-Euclidean Distance Measures in GeostatisticsMathematical Geology, 2006
- Spatial modelling using a new class of nonstationary covariance functionsEnvironmetrics, 2006
- Minimum dimension embedding of finite metric spacesJournal of Combinatorial Theory, Series A, 1986