Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 8 of 8 for “"Vecchia approximation"”.
-
Sparse Inverse Cholesky Factorization for Scalable Gaussian-Process Inference
… is computationally intractable and requires approximations, because decomposing the covariance matrix of the data vector requires cubic scaling computational complexity in the number of data points. A highly promising approach to achieve GP scalability is sparse inverse Cholesky (SIC) …
-
New Developments on Computer Experiments with Quantitative and Qualitative Inputs
… We extend existing methodologies, such as the Vecchia approximation, global-local Gaussian process, and local approximation Gaussian process, to accommodate both quantitative and qualitative inputs. Our proposed approach based on Vecchia approximation significantly improves computational …
-
Multifidelity methods for data fusion of environmental data
… developed. This framework leveraged the use of Vecchia approximation to make complex multifidelity models more efficient. Both the WMFGP and the scalable framework prove effective in modeling wind speed data from Lombardy, a region in northern Italy.
-
Deep Gaussian Process Surrogates for Computer Experiments
… bottlenecks. For this case I introduce the Vecchia approximation, popular for ordinary GPs in spatial data settings. I show that Vecchia-induced sparsity of Cholesky factors allows for linear computational scaling without compromising DGP accuracy or UQ. I vet both active learning and …
-
Fast Inference for Multi-scale and Global Spatial Processes
… structure, we propose a novel multi-scale Vecchia (MSV) approximation of GPs. In the MSV method, the increasingly small scales of spatial variation can be captured by increasingly large sets of variables, and then an accurate approximation of the spatial dependence is obtained from very …
-
Novel Machine Learning Algorithms for Large Spatial Data
… GP learning and prediction, hybrid covariance approximations that remain accurate across diverse dimensional and smoothness regimes, and modeling non-separable space-time dependence in real-world environmental applications. The first contribution introduces an iterative inference framework for …
-
EFFICIENT BAYESIAN MODELS FOR LARGE SPATIOTEMPORAL GAUSSIAN AND COUNT DATA
… by most previous models, which exhibit several approximation and posterior sampling challenges, we construct a spatiotemporal autoregressive gamma process guaranteed temporally stationary. We decompose latent Poisson variables to permit fully conjugate Gibbs sampling of the frailties and design …
-
Scalable Surrogates for Counts and Computer Experiments
… To address this gap, I propose a fully Bayesian, Vecchia-approximated, Poisson deep GP surrogate model. I demonstrate its improved predictive capability over competitors through multiple simulated examples. Further, I develop a novel, fully Bayesian framework for solving Bayesian inverse problems, …