AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Extending Mean-Field Variational Inference via Entropic Regularization: Theory and Computation

Bohan Wu, David M. Blei

JMLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Variational inference theory and gradients · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add entropic expressivity regularization (Xi-VI, small lambda) to mean-field VImean-field VI · Eight Schoolsposterior dependence between theta_1 and theta_7Improved
add Xi-VI instead of MFVI for max and min treatment effect credible intervalsMFVI, NFVI, SVGD, full-rank ADVI · Eight Schools95% credible interval width and coverageImproved
increase lambda beyond D (more regularization toward mean-field) (ablation)Xi-VI · Laplace linear regression (n=100, d=12)W2 error and runtimeMixed
increase lambda from below 1 to above 10 (ablation)Xi-VI · Eight Schoolsapproximation error vs runtimeMixed
increase number of pseudomarginal points N (ablation)Xi-VI · Laplace linear regression (lambda=10)approximation accuracyNo clear effect
use overdispersed pseudomarginals rather than underdispersed (ablation)Xi-VI one-step EOT correction · Eight Schoolssample dependency structureImproved
compare normalizing-flow VI with Xi-VIXi-VI · Eight SchoolsW2 and KL distanceNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy