Research map / Probabilistic and scientific modeling / Variational and generative sampling
Composing graphical models with neural networks for structured representations and fast inference
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 |
|---|---|---|---|
| natural gradient updates instead of standard gradient updates for the latent model parameters (ablation) | LDS SVAE on synthetic 1D dot image data · synthetic 1D dot bouncing image data | learning speed and robustness to step size | Helped |
| standard gradient updates at step sizes 0.1 and 0.05 instead of natural gradients (ablation) | LDS SVAE on synthetic 1D dot image data · synthetic 1D dot bouncing image data | stability of matrix parameters | Hurt |
Closest papers by meaning
- On Neural Networks as Infinite Tree-Structured Probabilistic Graphical Models (NeurIPS 2024 · similarity 0.88)
- Neural Variational Inference and Learning in Undirected Graphical Models (NeurIPS 2017 · similarity 0.88)
- Structured Recognition for Generative Models with Explaining Away (NeurIPS 2022 · similarity 0.87)
- Variational Message Passing with Structured Inference Networks (ICLR 2018 · similarity 0.87)
- Probabilistic Circuits for Variational Inference in Discrete Graphical Models (NeurIPS 2020 · similarity 0.86)
- Neural Snowflakes: Universal Latent Graph Inference via Trainable Latent Geometries (ICLR 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.