Research map / Architectures and efficiency / RNNs and normalization
Augmentable Gamma Belief Networks
Mingyuan Zhou, Yulai Cong, Bo Chen
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add four more layers (T=1 to T=5) to the PGBN (ablation) | PGBN K1max=128 · 20newsgroups multi-class (V=2000) | accuracy (%): 65.9 -> 67.5 | Improved |
| add four more layers (T=1 to T=5) to the PGBN (ablation) | PGBN K1max=512 · 20newsgroups multi-class (V=2000) | accuracy (%): 65.9 -> 69.2 | Improved |
| increase network depth with a limited first-layer width (ablation) | PGBN · 20newsgroups multi-class (V=33,420) | accuracy (%) | Improved |
| increase the upper bound of first-layer width with depth fixed (ablation) | PGBN · 20newsgroups multi-class (V=33,420) | accuracy (%) | Improved |
| add layers to a single-layer PGBN with K1max=200 (versus single-layer K1max=800) (ablation) | PGBN · 20newsgroups multi-class (V=33,420) | accuracy (%) | Improved |
| modify the PGBN from single-layer shallow to multilayer deep (ablation) | PGBN · 20newsgroups binary classification (ibm.pc.hardware vs mac.hardware; sci.electronics vs sci.med) | accuracy (%) | Improved |
| increase depth from 1 to 3 layers (ablation) | PGBN K1max=400 · 20newsgroups multi-class | seconds per training iteration: 5.6 -> 6.7 | Hurt |
| increase depth from 1 to 5 layers (ablation) | PGBN K1max=400 · 20newsgroups multi-class | seconds per training iteration: 5.6 -> 7.1 | Hurt |
| use PGBN T=5 K1max=128 instead of DocNADE with 128 features | PGBN · 20newsgroups multi-class (V=2000) | accuracy (%): 67.0 -> 67.5 | No clear effect |
| use PGBN T=5 K1max=512 instead of DocNADE with 512 features | PGBN · 20newsgroups multi-class (V=2000) | accuracy (%): 68.4 -> 69.2 | No clear effect |
| use normalized term frequencies instead of raw counts as logistic regression covariates | L2 logistic regression · 20newsgroups multi-class (V=33,420) | accuracy (%): 78.0 -> 79.4 | Improved |
| use all 61,188 terms instead of top 2000 terms excluding stopwords (raw counts) | L2 logistic regression · 20newsgroups multi-class | accuracy (%): 69.8 -> 75.8 | Improved |
Closest papers by meaning
- A Non-Negative Deep VAE: The Generalized Gamma Belief Network (TPAMI 2026 · similarity 0.91)
- Multimodal Poisson Gamma Belief Network (AAAI 2018 · similarity 0.86)
- Sparse Poisson Gamma Belief Networks for High-Dimensional Sparse Count Data (AAAI 2026 · similarity 0.86)
- Convolutional Poisson Gamma Belief Network (ICML 2019 · similarity 0.85)
- GAMMA: Gated Multi-hop Message Passing for Homophily-Agnostic Node Representation in GNNs (NeurIPS 2025 · similarity 0.85)
- Learning in Deep Factor Graphs with Gaussian Belief Propagation (ICML 2024 · similarity 0.83)
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.
