AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Online Bayesian Passive-Aggressive Learning

Tianlin Shi, Jun Zhu

JMLR 2017 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning

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 online BayesPA mini-batch updates instead of batch MedLDA trainingMedLDA / Gibbs MedLDA · 20NGtraining time: more than an order of magnitude fasterImproved
use online BayesPA instead of batch multi-task MedLDAMedLDAmt · Wiki 1.1M documents (multi-task)training time at comparable F1: about two orders of magnitude fasterImproved
add per-mini-batch supervision (BayesPA) versus unsupervised online spLDA+SVMonline topic model · 20NG (K=80)test error convergenceImproved
use online BayesPA instead of batch trainingpaMedLDA · 20NGaccuracyNo clear effect
use medium mini-batch size (64 or 256) rather than very small or large (ablation)paMedLDA (K=40) · 20NGconvergence speed: |B| = 64 or 256 converge fasterImproved
increase optimization rounds I from 1 to 2 (ablation)paMedLDAave and paMedHDPave · 20NGaccuracy: I = 1 suffers largelyImproved
increase optimization rounds I (ablation)paMedLDAgibbs and paMedHDPgibbs · 20NGaccuracyNo clear effect
increase number of local samples J (ablation)BayesPA topic models · 20NGaccuracyNo clear effect
increase local samples J from 1 to 5 (beta=0) (ablation)paMedLDAgibbs · 20NG (K=40)accuracy: 0.783 -> 0.808Improved
add burn-in steps with number of kept samples fixed (ablation)paMedLDA · 20NG (K=40)accuracy: (J, beta) = (3, 0), (5, 2), (9, 6) yield similar resultsNo clear effect
use averaging classifier instead of Gibbs classifier (ablation)paMedLDA · 20NGtraining speedImproved
increase number of topics with one-pass training (ablation)online Gibbs MedLDA · 20NG alt.atheism vs talk.religion.miscaccuracyHurt

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