Research map / Graph learning / Graph self-supervised learning
Graph-based Isometry Invariant Representation Learning
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 TIGraNet's graph-based transformation-invariant architecture instead of ConvNet | Image classification without transformed training examples · MNIST-rot | accuracy: 44.3 -> 83.8 | Helped |
| Use TIGraNet's graph-based transformation-invariant architecture instead of STN | Image classification without transformed training examples · MNIST-rot | accuracy: 44.5 -> 83.8 | Helped |
| Use TIGraNet's graph-based transformation-invariant architecture instead of ConvNet | Image classification without transformed training examples · MNIST-trans | accuracy: 43.5 -> 79.6 | Helped |
| Use TIGraNet's graph-based transformation-invariant architecture instead of STN | Image classification without transformed training examples · MNIST-trans | accuracy: 67.1 -> 79.6 | Helped |
| Use TIGraNet instead of HarmNet | Object classification from different viewpoints · ETH-80 | accuracy: 94.0 -> 95.1 | Helped |
Closest papers by meaning
- Isometric Transformation Invariant and Equivariant Graph Convolutional Networks (ICLR 2021 · similarity 0.90)
- GraphTER: Unsupervised Learning of Graph Transformation Equivariant Representations via Auto-Encoding Node-Wise Transformations (CVPR 2020 · similarity 0.87)
- Equivariant Machine Learning on Graphs with Nonlinear Spectral Filters (NeurIPS 2024 · similarity 0.87)
- Neural Isometries: Taming Transformations for Equivariant ML (NeurIPS 2024 · similarity 0.86)
- On Making $SE(2)$-Invariant Networks Optimal (NeurIPS 2026 · similarity 0.86)
- Unitary Convolutions for Learning on Graphs and Groups (NeurIPS 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.