AI Paper Atlas

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

Decoupling Sparsity and Smoothness in the Dirichlet Variational Autoencoder Topic Model

Sophie Burkhardt, Stefan Kramer

JMLR 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Structured latent VAEs · 58 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
decouple sparsity and smoothness (DVAE Sparse vs DVAE)DVAE · 20news (prior 0.02, 50 topics)perplexity: 5009 -> 1066Improved
decouple sparsity and smoothness (DVAE Sparse vs DVAE)DVAE · 20news (prior 0.02, 50 topics)topic coherence: 0.324 -> 0.248Hurt
sample the KL-divergence during training instead of analytical (ablation)DVAE · 20news (prior 0.02)topic coherence: 0.33 -> 0.35Improved
sample the KL-divergence during training instead of analytical (ablation)DVAE · 20news (prior 0.02)final perplexity and KLImproved
remove batch normalization (ablation)Dirichlet VAEs · topic modelstopic qualityHurt
omit gradient correction term when shape augmentation is used (ablation)DVAE · topic modelsperformanceNo clear effect
RSVI Dirichlet reparameterization vs Laplace approximation (DVAE vs ProdLDA)DVAE · 20news (prior 0.02, 50 topics)topic coherence: 0.223 -> 0.324Improved
RSVI vs Weibull reparameterizationDVAE · 20news (prior 0.02, 50 topics)topic coherence: 0.271 -> 0.324Improved
increase Dirichlet prior from 0.02 to 0.6DVAE · average coherence rankaverage rank: 1.125 -> 1.750Hurt
Dirichlet prior (DVAE) vs Gaussian prior (NVDM)topic model · 20news (prior 0.02, 50 topics)topic coherence: 0.140 -> 0.324Improved
Dirichlet prior (DVAE) vs Gaussian prior (NVDM)topic model · all datasets and topic counts (prior 0.02)perplexity average rank: 7.125 vs 1.000Hurt

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