Research map / Probabilistic and scientific modeling / Variational and generative sampling
Convolutional Poisson Gamma Belief Network
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 CPGBN instead of the strongest listed unsupervised baseline | unsupervised feature extraction · MR | accuracy: 57.0$\pm0.5$ -> 63.6$\pm0.8$ | Helped |
| use CPGBN instead of the strongest listed unsupervised baseline | unsupervised feature extraction · TREC | accuracy: 67.9$\pm1.5$ -> 74.4$\pm0.6$ | Helped |
| use CPGBN instead of the strongest listed unsupervised baseline | unsupervised feature extraction · SUBJ | accuracy: 78.3$\pm1.2$ -> 81.5$\pm0.6$ | Helped |
| use three stochastic hidden layers instead of one (ablation) | sCPGBN · SUBJ | accuracy: 93.4$\pm0.1$ -> 93.8$\pm0.1$ | Helped |
| use three stochastic hidden layers instead of one (ablation) | sCPGBN · ELEC | accuracy: 91.6$\pm0.3$ -> 92.2$\pm0.2$ | Helped |
| use three stochastic hidden layers instead of one (ablation) | sCPGBN · IMDB | accuracy: 91.8$\pm0.3$ -> 92.6$\pm0.2$ | Helped |
| increase the network from one to two hidden layers (ablation) | CPGBN · MR | accuracy: 61.5$\pm0.8$ -> 62.4$\pm0.7$ | Helped |
| increase the network from one to two hidden layers (ablation) | CPGBN · TREC | accuracy: 68.4$\pm0.8$ -> 73.4$\pm0.8$ | Helped |
| increase the network from one to two hidden layers (ablation) | CPGBN · SUBJ | accuracy: 77.4$\pm0.8$ -> 81.2$\pm0.8$ | Helped |
| vary the convolution filter width from 1 to 2 (ablation) | CPFA · SUBJ | accuracy: $74.9 \pm 0.9$ -> $77.3 \pm 0.4$ | Helped |
| vary the convolution filter width from 1 to 3 (ablation) | CPFA · SUBJ | accuracy: $74.9 \pm 0.9$ -> $77.5 \pm 0.5$ | Helped |
| vary the convolution filter width from 1 to 4 (ablation) | CPFA · SUBJ | accuracy: $74.9 \pm 0.9$ -> $77.8 \pm 0.4$ | Helped |
Closest papers by meaning
- Bidirectional Convolutional Poisson Gamma Dynamical Systems (NeurIPS 2020 · similarity 0.92)
- Deep Relational Topic Modeling via Graph Poisson Gamma Belief Network (NeurIPS 2020 · similarity 0.91)
- Sawtooth Factorial Topic Embeddings Guided Gamma Belief Network (ICML 2021 · similarity 0.88)
- Multimodal Poisson Gamma Belief Network (AAAI 2018 · similarity 0.87)
- Deep Latent Dirichlet Allocation with Topic-Layer-Adaptive Stochastic Gradient Riemannian MCMC (ICML 2017 · similarity 0.86)
- Poisson-Randomised DirBN: Large Mutation is Needed in Dirichlet Belief Networks (ICML 2021 · 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.