Application of SGRLD to Large-Scale Ocean Temperature Data: The Argo Case Study
1 Jun 2025
SGRLD on Argo ocean data: 4x lower MSE, 20x faster than NNGP, with 93% prediction interval coverage on 32K temperature measurements.
Battle of the Algorithms: Why SGRLD Beats the Competition in GP Inference
1 Jun 2025
SGRLD outperforms competing SGMCMC methods and NNGP in MSE and ESS per minute, showing superior parameter estimation and computational efficiency.
New Algorithm Makes Complex Spatial Models Lightning Fast with Vecchia Magic
31 May 2025
Matern GP model with Vecchia approximation enables linear-time computation and unbiased gradient estimation for scalable SGMCMC spatial inference.
Breaking the iid Barrier: How Stochastic Gradients Are Going Spatial
31 May 2025
SG methods scale ML but struggle with correlated spatial data. Authors extend SGLD using Vecchia approximation for unbiased GP gradient estimates.
How Spatial Scientists Are Solving the Gaussian Process Scaling Challenge
31 May 2025
Reviews GP scaling methods: low-rank, tapering, divide-conquer, Vecchia. All reduce cubic to linear cost, but MCMC still needs thousands of iterations.
Stochastic Gradient MCMC for Large-Scale Gaussian Process Spatial Modeling
31 May 2025
SGMCMC + Vecchia approximation enables scalable Bayesian GP inference for large spatial datasets, achieving linear time complexity and competitive performance.