AI Paper Atlas

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

Gradient Estimation for Mixture Variational Inference

Javier Burroni, Daniel Sheldon

JMLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Variational inference theory and gradients

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
use reparameterization-based estimator (J&K) instead of score functionmixture VI · Deep Markov model K=2test ELBO: −11.26 -> −7.72Improved
use post-stratified instead of score functionmixture VI · Deep Markov model K=5test ELBO: −11.24 -> −11.20Improved
use J&K or implicit reparameterization instead of post-stratifiedmixture VI · Deep Markov model K=2test ELBO: −10.64 -> −7.72Improved
use a mixture (K>1) instead of single Gaussianvariational posterior · Deep Markov modeltest ELBO: −7.61 -> −7.75Hurt
add Sticking-the-Landing to J&K (ablation)J&K estimator · Deep Markov model K=2ELBO: +0.16Improved
add Sticking-the-Landing to post-stratified (ablation)post-stratified estimator · Deep Markov model K=3ELBO: +0.01No clear effect
add Sticking-the-Landing (ablation)estimators with single particle · MNIST-like datasets, diagonal GaussiansELBO: 24 of 32 configurations (75 %)Mixed
add Sticking-the-Landing (ablation)stratified, post-stratified, J&K · Baseball K=5 single particletest ELBOHurt
use implicit reparameterization instead of score functionmixture VI · Deep Markov model running timerunning time increase (%): 64 % to 68 % longerHurt
use J&K instead of score functionmixture VI · Deep Markov model running timerunning time increase (%): 20 % longerHurt
use stratified instead of single-sample estimatorsmixture VI · BaseballELBOImproved
use stratified estimator in chains of mixturesstratified estimator · Deep Markov modelfeasibilityHurt

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