Back to the program

Contributed Papers

Novel Approaches to the Scalability Challenges of Spatial and Space-Time Inference

Wed, Aug 5, 2:00 PM - 3:50 PM Room CC-105 Thomas M. Menino Convention & Exhibition Center

Drew YargerChair

Section on Statistics and the Environment

7 Presentations

A Multi-Stage Approach to Fit Marked Point Process Models with Applications to Environmental Science

2:05 PM - 2:20 PM

Rachael Ren (The University of Texas At Austin)

Co-authors: Mevin Hooten (The University of Texas at Austin), Toryn Schafer (Texas A&M University), Benjamin Hoose (Texas A&M University), Jamie Womble (National Park Service), Scott Gende (National Park Service)

spatial point process recursive Bayes MCMC parallel computing species distribution modeling presence-only data

A Scalable Variational Approach for Modeling Multivariate Spatial Data with Variational Inference Normalizing Flows

2:20 PM - 2:35 PM

Jin Hyung Lee (Purdue University)

Co-authors: Anindya Bhadra (Purdue University)

Variational Inference Multivariate Spatial Process Bayesian Inference Normalizing Flows Spatial Statistics Scalable Method

Canonical Polyadic Core Covariance Estimation for Matrix-Variate Data

2:35 PM - 2:50 PM

Bongjung Sung (Duke University)

Co-authors: Peter Hoff (Duke University)

Canonical polyadic tensor decomposition Core covariance matrix Kronecker-core decomposition Low-rank Separable covariance matrix

Efficient Generative Modeling of Non-Gaussian Global Climate Fields via Composite Transformations

2:50 PM - 3:05 PM

Johannes Brachem

Co-authors: Paul Wiemann (The Ohio State University), Matthias Katzfuss (University of Wisconsin–Madison)

Emulating internal climate variability Stochastic surrogate Gaussian processes Non-Gaussian spatial field Transformation model

Improving Bayesian computation of latent nearest neighbor Gaussian processes for generalized spatial models

3:05 PM - 3:20 PM

Wilson Wright (University of Missouri)

Co-authors: Mevin Hooten (The University of Texas at Austin)

Geostatistics Hamiltonian Monte Carlo Markov chain Monte Carlo recursive Bayesian computation spatial statistics

Scalable Spatial Stream Network (S3N) Models

3:20 PM - 3:35 PM

Jessica Kunke (Montana State University)

Co-authors: Julian D. Olden (University of Washington), Tyler McCormick (University of Washington)

spacetime processes ecology networks computational efficiency SSNs statistical computing

Sparse Autoregressive Flow-Based Variational Inference for Non-Gaussian Latent Fields

3:35 PM - 3:50 PM

Paul Wiemann (The Ohio State University)

Amortized variational inference Latent-field model Spatial Statistics Bayesian Statistics Normalizing Flow

↑