A1409
Title: Stochastic gradient MCMC for massive geostatistical data
Authors: Mohamed Abba - Amazon (United States)
Brian Reich - North Carolina State University (United States)
Reetam Majumder - University of Arkansas (United States) [presenting]
Brandon Feng - North Carolina State University (United States)
Abstract: Gaussian processes (GPs) are commonly used for prediction and inference for spatial data analyses. However, since estimation and prediction tasks have cubic time and quadratic memory complexity in the number of locations, GPs are difficult to scale to large spatial datasets. The Vecchia approximation induces sparsity in the dependence structure and is one of several methods proposed to scale GP inference. The purpose is to add to the substantial research in this area by developing a stochastic gradient Markov chain Monte Carlo (SGMCMC) framework for efficient computation in GPs. At each step, the algorithm subsamples a minibatch of locations and subsequently updates process parameters through a Vecchia-approximated GP likelihood. Since the Vecchia-approximated GP has a time complexity that is linear in the number of locations, this results in scalable estimation in GPs. Through simulation studies, SGMCMC is demonstrated to be competitive with state-of-the-art scalable GP algorithms in terms of computational time and parameter estimation. An application of the method is also provided using the Argo dataset of ocean temperature measurements.
