Research map / Probabilistic and scientific modeling / Variational and generative sampling
Principal Component Flows
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 |
|---|---|---|---|
| add orthogonality regularizer (IP=0) to train principal component flows | normalizing flow on 2D synthetic datasets · 2D synthetic datasets (8 datasets) | test log likelihood: differences of about 0.01-0.02 nats | No clear effect |
| add orthogonality regularizer (IP=0) to train principal component flows | normalizing flow on 2D synthetic datasets · 2D synthetic datasets (8 datasets) | pointwise mutual information IP: IP drops from 0.39-1.60 to 0.00 | Helped |
| train injective flows with the iPCF objective instead of injective change of variables | injective normalizing flow on MNIST · MNIST (10-dim latent) | training cost: roughly 10 times less resources | Helped |
| use manifold corrected density (Definition 7) to select contours | PCF trained on variable-rank manifold data · variable-dimensional manifold dataset | forward KL divergence to true density | Helped |
Closest papers by meaning
- Injective flows for star-like manifolds (ICLR 2025 · similarity 0.87)
- Lifting Architectural Constraints of Injective Flows (ICLR 2024 · similarity 0.86)
- Kernelised Normalising Flows (ICLR 2024 · similarity 0.86)
- CONTRA: Conformal Prediction Region via Normalizing Flow Transformation (ICLR 2025 · similarity 0.86)
- Normalizing Flows are Capable Generative Models (ICML 2025 · similarity 0.86)
- Structured Coupling for Flow Matching (NeurIPS 2026 · 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.