Research map / Graph learning / Graph self-supervised learning
Matrix Completion for Graph-Based Deep Semi-Supervised 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 |
|---|---|---|---|
| add triplet regularization loss on the matrix-completion graph (ablation) | GSCNN · MNIST (100 labels) | error rate (%): 1.14 -> 0.84 | Helped |
| add triplet regularization loss (ablation) | GSCNN · SVHN (1000 labels) | error rate (%): 8.41 -> 5.13 | Helped |
| add triplet regularization loss (ablation) | GSCNN · NORB (1000 labels) | error rate (%): 8.71 -> 7.01 | Helped |
| add triplet regularization loss (ablation) | GSCNN · CIFAR-10 (4000 labels) | error rate (%): 18.03 -> 15.49 | Helped |
| train CNN with graph triplet loss instead of matrix completion label prediction alone (ablation) | GSCNN · SVHN (1000 labels) | error rate (%): 10.06 -> 5.13 | Helped |
| increase labeled samples per class from 100 to 800 (ablation) | GSCNN · NORB | error rate (%): 9.88 -> 5.07 | Helped |
| increase labeled samples per class from 100 to 800 (ablation) | GSCNN · CIFAR-10 | error rate (%): 18.98 -> 14.51 | Helped |
| GSCNN vs best prior SSL method | GSCNN · MNIST (100 labels) | error rate (%): 0.91 -> 0.84 | Helped |
| GSCNN vs best prior SSL method | GSCNN · SVHN (1000 labels) | error rate (%): 5.43 -> 5.13 | Helped |
| GSCNN vs best prior SSL method | GSCNN · CIFAR-10 (4000 labels) | error rate (%): 16.55 -> 15.49 | Helped |
Closest papers by meaning
- Contrastive and Generative Graph Convolutional Networks for Graph-based Semi-Supervised Learning (AAAI 2021 · similarity 0.91)
- Semi-Supervised Learning With Graph Learning-Convolutional Networks (CVPR 2019 · similarity 0.90)
- Label Efficient Semi-Supervised Learning via Graph Filtering (CVPR 2019 · similarity 0.89)
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.89)
- Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification (AAAI 2024 · similarity 0.88)
- Normalize Then Propagate: Efficient Homophilous Regularization for Few-Shot Semi-Supervised Node Classification (AAAI 2025 · similarity 0.87)
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.