Research map / Probabilistic and scientific modeling / Variational and generative sampling
Online Bayesian Moment Matching for Topic Modeling with Unknown Number of Topics
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use uniform alpha for online HDP instead of learned alpha | online HDP · Reuters/NIPS/Yelp | per-word test log likelihood | Hurt |
| Replace basic fixed-T moment matching with DDM/TDM that infer number of topics | basic moment matching · NIPS and Yelp | per-word test log likelihood | Helped |
| Initialize beta_{t,w} to 1/sqrt(V) instead of 1 in TDM/DDM (ablation) | DDM and TDM on synthetic data · synthetic data | estimated number of topics | Hurt |
| Use exponentially decreasing prior on T instead of uniform prior (ablation) | DDM and TDM on synthetic data · synthetic data | estimated number of topics vs true T | Helped |
Closest papers by meaning
- A Reduction for Efficient LDA Topic Reconstruction (NeurIPS 2018 · similarity 0.87)
- Diagnosing and Improving Topic Models by Analyzing Posterior Variability (AAAI 2018 · similarity 0.87)
- Discriminative Topic Modeling with Logistic LDA (NeurIPS 2019 · similarity 0.87)
- Temporal Topic Analysis with Endogenous and Exogenous Processes (AAAI 2016 · similarity 0.86)
- E-LDA: Toward Interpretable LDA Topic Models with Strong Guarantees in Logarithmic Parallel Time (ICML 2025 · similarity 0.86)
- On the Affinity, Rationality, and Diversity of Hierarchical Topic Modeling (AAAI 2024 · similarity 0.85)
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.