AI Paper Atlas

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

Pólya Urn Latent Dirichlet Allocation: A Doubly Sparse Massively Parallel Sampler

Alexander Terenin, Måns Magnusson, Leif Jonsson, David Draper

TPAMI 2018 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Tractable probabilistic models · 24 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
replace Dirichlet sampling of Phi with Poisson Polya urnpartially collapsed LDA · Enron/NYT/PubMed, K=100,1000runtimeImproved
replace Dirichlet sampling of Phi with Poisson Polya urnpartially collapsed LDA · Enron/NYT/PubMed, per iterationlog-posteriorNo clear effect
replace Dirichlet with Poisson Polya urn at K=10partially collapsed LDA · Enron/NYT/PubMed K=10runtime and log-posteriorNo clear effect
sample Phi with alias tables and Gaussian approximationPhi sampling · Enron/NYT/PubMedPhi sampling time: factor of four to eightImproved
increase topics K from 100 to 1000 with Polya urnz sampling · PubMedz sampling time: by as much as 20%Improved
use Polya Urn LDA instead of partially collapsed, sparse and light AD-LDAparallel LDA samplers · NYT, K=100 and K=1000runtimeImproved
use Polya Urn LDA instead of partially collapsed LDApartially collapsed LDA · NYT held-outtest log-likelihood and topic coherenceNo clear effect
use Polya Urn LDA on single core vs sparse fully collapsedfully collapsed sparse LDA · single coreruntimeImproved
increase vocabulary size / lower rare word threshold (ablation)Polya Urn LDA vs partially collapsed · NYTrelative runtimeImproved
add more CPU coresPolya Urn LDA · parallel scalingruntimeImproved

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