Research map / Probabilistic and scientific modeling / Variational and generative sampling
Topic Modeling via Full Dependence Mixtures
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 FDM instead of SparseLDA with the true hyperparameter | Topic modeling on the synthetic corpus · Synthetic corpus | average matching ℓ1 distance: 0.41 ± 0.0004 -> 0.66 ± 0.005 | Hurt |
| Use FDM instead of SparseLDA with the modified hyperparameter | Topic modeling on the synthetic corpus · Synthetic corpus | average matching ℓ1 distance: 0.66 ± 0.01 -> 0.66 ± 0.005 | No clear effect |
| Use FDM instead of EP | Topic modeling on the synthetic corpus · Synthetic corpus | average matching ℓ1 distance: 1.05 ± 0.005 -> 0.66 ± 0.005 | Helped |
| Use FDM instead of SparseLDA | Topic modeling · NeurIPS | test-set average log-likelihood: -5.57 -> -5.62 | No clear effect |
| Use FDM instead of EP | Topic modeling · NeurIPS | test-set average log-likelihood: -6.31 -> -5.62 | Helped |
| Use FDM instead of SparseLDA | Topic modeling · Twitter | test-set average log-likelihood: -5.70 -> -5.26 | Helped |
| Use FDM instead of EP | Topic modeling · Twitter | test-set average log-likelihood: -8.5 -> -5.26 | Helped |
Closest papers by meaning
- Representing Mixtures of Word Embeddings with Mixtures of Topic Embeddings (ICLR 2022 · similarity 0.87)
- Algorithms for Generalized Topic Modeling (AAAI 2018 · similarity 0.86)
- textTOvec: DEEP CONTEXTUALIZED NEURAL AUTOREGRESSIVE TOPIC MODELS OF LANGUAGE WITH DISTRIBUTED COMPOSITIONAL PRIOR (ICLR 2019 · similarity 0.85)
- Online Bayesian Moment Matching for Topic Modeling with Unknown Number of Topics (NeurIPS 2016 · similarity 0.85)
- Discovering Discrete Latent Topics with Neural Variational Inference (ICML 2017 · similarity 0.85)
- Dirichlet Multinomial Mixture with Variational Manifold Regularization: Topic Modeling over Short Texts (AAAI 2019 · 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.