Projective Integral Updates for High-Dimensional Variational Inference

DOI10.1137/22m1529919arXiv2301.08374OpenAlexW4391645332WikidataQ128381766 ScholiaQ128381766MaRDI QIDQ6131419

Jed A. Duersch

Publication date: 5 April 2024

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2301.08374

zbMATH Keywords

cubaturequadraturequasi-Newtonvariational inferencespike-and-slabHessian approximationHadamard basisGaussian mean-field

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Parametric inference under constraints (62F30) Bayesian inference (62F15) Monte Carlo methods (65C05) Numerical optimization and variational techniques (65K10) Numerical quadrature and cubature formulas (65D32)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Numerical integration using sparse grids
• Smolyak cubature of given polynomial degree with few nodes for increasing dimension
• Simple cubature formulas with high polynomial exactness
• Generalized parallel tempering on Bayesian inverse problems
• An estimator for the diagonal of a matrix
• Construction of fully symmetric numerical integration formulas
• A Systematization of the Unscented Kalman Filter Theory
• Exactness of Quadrature Formulas
• BAYESIAN INFERENCE OF STOCHASTIC REACTION NETWORKS USING MULTIFIDELITY SEQUENTIAL TEMPERED MARKOV CHAIN MONTE CARLO
• Cubature, Approximation, and Isotropy in the Hypercube
• High-dimensional integration: The quasi-Monte Carlo way
• Algorithms for Numerical Analysis in High Dimensions
• On Information and Sufficiency
• A Stochastic Approximation Method