Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386080908> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4386080908 abstract "Bayesian Experimental Design (BED), which aims to find the optimal experimental conditions for Bayesian inference, is usually posed as to optimize the expected information gain (EIG). The gradient information is often needed for efficient EIG optimization, and as a result the ability to estimate the gradient of EIG is essential for BED problems. The primary goal of this work is to develop methods for estimating the gradient of EIG, which, combined with the stochastic gradient descent algorithms, result in efficient optimization of EIG. Specifically, we first introduce a posterior expected representation of the EIG gradient with respect to the design variables. Based on this, we propose two methods for estimating the EIG gradient, UEEG-MCMC that leverages posterior samples generated through Markov Chain Monte Carlo (MCMC) to estimate the EIG gradient, and BEEG-AP that focuses on achieving high simulation efficiency by repeatedly using parameter samples. Theoretical analysis and numerical studies illustrate that UEEG-MCMC is robust agains the actual EIG value, while BEEG-AP is more efficient when the EIG value to be optimized is small. Moreover, both methods show superior performance compared to several popular benchmarks in our numerical experiments." @default.
- W4386080908 created "2023-08-23" @default.
- W4386080908 creator A5026669579 @default.
- W4386080908 creator A5066997973 @default.
- W4386080908 date "2023-08-18" @default.
- W4386080908 modified "2023-09-27" @default.
- W4386080908 title "On Estimating the Gradient of the Expected Information Gain in Bayesian Experimental Design" @default.
- W4386080908 doi "https://doi.org/10.48550/arxiv.2308.09888" @default.
- W4386080908 hasPublicationYear "2023" @default.
- W4386080908 type Work @default.
- W4386080908 citedByCount "0" @default.
- W4386080908 crossrefType "posted-content" @default.
- W4386080908 hasAuthorship W4386080908A5026669579 @default.
- W4386080908 hasAuthorship W4386080908A5066997973 @default.
- W4386080908 hasBestOaLocation W43860809081 @default.
- W4386080908 hasConcept C107673813 @default.
- W4386080908 hasConcept C111350023 @default.
- W4386080908 hasConcept C11413529 @default.
- W4386080908 hasConcept C126255220 @default.
- W4386080908 hasConcept C153258448 @default.
- W4386080908 hasConcept C154945302 @default.
- W4386080908 hasConcept C160234255 @default.
- W4386080908 hasConcept C17744445 @default.
- W4386080908 hasConcept C199539241 @default.
- W4386080908 hasConcept C2776214188 @default.
- W4386080908 hasConcept C2776359362 @default.
- W4386080908 hasConcept C33923547 @default.
- W4386080908 hasConcept C41008148 @default.
- W4386080908 hasConcept C50644808 @default.
- W4386080908 hasConcept C94625758 @default.
- W4386080908 hasConceptScore W4386080908C107673813 @default.
- W4386080908 hasConceptScore W4386080908C111350023 @default.
- W4386080908 hasConceptScore W4386080908C11413529 @default.
- W4386080908 hasConceptScore W4386080908C126255220 @default.
- W4386080908 hasConceptScore W4386080908C153258448 @default.
- W4386080908 hasConceptScore W4386080908C154945302 @default.
- W4386080908 hasConceptScore W4386080908C160234255 @default.
- W4386080908 hasConceptScore W4386080908C17744445 @default.
- W4386080908 hasConceptScore W4386080908C199539241 @default.
- W4386080908 hasConceptScore W4386080908C2776214188 @default.
- W4386080908 hasConceptScore W4386080908C2776359362 @default.
- W4386080908 hasConceptScore W4386080908C33923547 @default.
- W4386080908 hasConceptScore W4386080908C41008148 @default.
- W4386080908 hasConceptScore W4386080908C50644808 @default.
- W4386080908 hasConceptScore W4386080908C94625758 @default.
- W4386080908 hasLocation W43860809081 @default.
- W4386080908 hasOpenAccess W4386080908 @default.
- W4386080908 hasPrimaryLocation W43860809081 @default.
- W4386080908 hasRelatedWork W2027276540 @default.
- W4386080908 hasRelatedWork W2146827233 @default.
- W4386080908 hasRelatedWork W2464065341 @default.
- W4386080908 hasRelatedWork W2753218748 @default.
- W4386080908 hasRelatedWork W2774409638 @default.
- W4386080908 hasRelatedWork W3165662366 @default.
- W4386080908 hasRelatedWork W4311165627 @default.
- W4386080908 hasRelatedWork W4318483226 @default.
- W4386080908 hasRelatedWork W4364383560 @default.
- W4386080908 hasRelatedWork W4377141721 @default.
- W4386080908 isParatext "false" @default.
- W4386080908 isRetracted "false" @default.
- W4386080908 workType "article" @default.