Efficient estimation of expected information gain in Bayesian experimental design with multi-index Monte Carlo

DOI10.1007/S11222-024-10522-5MaRDI QIDQ6657804zbMATH OpenFDO

Authors Xinting Du, Hejin Wang

Publication date 7 January 2025

Published in Statistics and Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/S11222-024-10522-5

zbMATH Keywords

Bayesian experimental designexpected information gainmulti-index Monte Carlo

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Optimal statistical designs (62K05) Monte Carlo methods (65C05)

Recommendations

• Multilevel Monte Carlo estimation of expected information gains
• Multilevel double loop Monte Carlo and stochastic collocation methods with importance sampling for Bayesian optimal experimental design
• Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
• Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs
• Fast estimation of expected information gains for Bayesian experimental designs based on Laplace approximations

Cites work

• A Mathematical Theory of Communication
• Bayesian experimental design: A review
• Convergence analysis of multilevel Monte Carlo variance estimators and application for random obstacle problems
• Decision-making under uncertainty: using MLMC for efficient estimation of EVPPI
• Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
• Implementation and analysis of an adaptive multilevel Monte Carlo algorithm
• Multi-index Monte Carlo: when sparsity meets sampling
• Multilevel and multi-index Monte Carlo methods for the McKean-Vlasov equation
• Multilevel double loop Monte Carlo and stochastic collocation methods with importance sampling for Bayesian optimal experimental design
• Multilevel Monte Carlo estimation of expected information gains
• Multilevel Monte Carlo methods and applications to elliptic PDEs with random coefficients
• Multilevel Monte Carlo Path Simulation
• On a Measure of the Information Provided by an Experiment
• On Information and Sufficiency
• Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs

Cited in

(2)

• Multilevel double loop Monte Carlo and stochastic collocation methods with importance sampling for Bayesian optimal experimental design
• Bayesian optimal design accelerates discovery of soft material properties from bubble dynamics

This page was built for publication: Efficient estimation of expected information gain in Bayesian experimental design with multi-index Monte Carlo

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6657804)