AI Paper Atlas

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

Online but Accurate Inference for Latent Variable Models with Local Gibbs Sampling

Christophe Dupuy, Francis Bach

JMLR 2017 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Variational methods for sequence models · 56 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
use Gibbs local sampling in online EM (G-OEM) instead of variational online LDAG-OEM · LDA on six datasetsheld-out log-perplexityImproved
use frequentist online EM instead of variational Bayesian (OLDA, SVB)G-OEM, V-OEM++, SPLDA · LDA on all datasetsheld-out log-perplexityImproved
use Gibbs sampling instead of variational inference for local updatesG-OEM · LDA, all experimentsheld-out log-perplexityImproved
add inference boosting to online EM (ablation)V-OEM++ · LDA, Synthetic and IMDBlog-perplexityImproved
add inference boosting to online EM (ablation)G-OEM++ · LDAlog-perplexityNo clear effect
use step-size 1/sqrt(t) instead of 1/t (ablation)V-OEM++ vs SPLDA · LDAlog-perplexityImproved
average iterates for kappa below 0.5 (ablation)G-OEM · Synthetic and IMDBperplexityImproved
average iterates for kappa above 0.5 (ablation)G-OEM · Synthetic and IMDBperplexityHurt
increase internal updates from P=20 (1 pass) to P=200 (4 passes) (ablation)OLDA · IMDBlog-perplexity: 681.9 -> 682.6No clear effect
increase internal updates from P=20 (1 pass) to P=200 (4 passes) (ablation)SVB · IMDBlog-perplexity: 684.5 -> 683.8No clear effect
use fixed point update for alpha instead of gamma prior (ablation)G-OEM · Synthetic, K=10perplexityImproved
use G-OEM instead of VarGibbsG-OEM · LDA and HDPperplexityImproved

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