Posterior Inference for Sparse Hierarchical Non-stationary Models
Gaussian processes are valuable tools for non-parametric modelling, where typically an assumption of stationarity is employed. While removing this assumption can improve prediction, fitting such models is challenging. In this work, hierarchical models are constructed based on Gaussian Markov random fields with stochastic spatially varying parameters. Importantly, this allows for non-stationarity while also addressing the computational burden through a sparse representation of the precision matrix. The prior field is chosen to be Matérn, and two hyperpriors, for the spatially varying parameters, are considered. One hyperprior is Ornstein-Uhlenbeck, formulated through an autoregressive AR(1) process. The other corresponds to the widely used squared exponential. Furthermore, all hyperparameters are treated as unknowns to achieve full uncertainty quantification. In this setting, efficient Markov chain Monte Carlo (MCMC) sampling is challenging due to the strong coupling a posteriori of the parameters and hyperparameters. We develop and compare three MCMC schemes, which are adaptive and therefore free of parameter tuning. An assessment of the ability of the methods to efficiently explore the posterior distribution and to account for non-stationarity is presented. Experiments in both simulated data and a real-world problem of cosmic radio noise are performed. We empirically consider posterior consistency of the estimates with respect to the discretisation scheme and explore the efficiency of the sampling methodologies under different settings.
READ FULL TEXT