Research map / Graph learning / Graph self-supervised learning
Graph Convolutional Gaussian Processes
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 graph-weighted geodesic patches with a shared GP patch response instead of a standard convolutional GP | GCGP · MNIST | error rate: 2.1% -> 1.7% | Helped |
| Use graph-weighted geodesic patches with a shared GP patch response instead of a full-input RBF GP | GCGP · MNIST | error rate: 1.9% -> 1.7% | Helped |
| Use spatial graph convolutions instead of spectral graph filters | GCGP · MNIST Superpixel 75 | error rate: 24.4% -> 4.2% | Helped |
| Use GCGP instead of MoNet | GCGP · MNIST Superpixel 75 | error rate: 8.9% -> 4.2% | Helped |
| Reduce training data to 500 examples per class (ablation) | GCGP · MNIST Superpixel 75 | error rate: 4.2% -> 8.3% | Hurt |
| Reduce training data to 100 examples per class (ablation) | GCGP · MNIST Superpixel 75 | error rate: 4.2% -> 13.7% | Hurt |
| Reduce training data to 1000 examples per class (ablation) | GCGP · MNIST Superpixel 75 | error rate: 4.2% -> 6.3% | Hurt |
| Use GCGP instead of MoNet | GCGP · MPI Faust, 500 vertices | error rate: 40.00% -> 23.33% | Helped |
| Use GCGP instead of MoNet | GCGP · MPI Faust, 1000 vertices | error rate: 33.33% -> 10.00% | Helped |
| Use GCGP instead of MoNet | GCGP · MPI Faust, 2500 vertices | error rate: 33.33% -> 3.33% | Helped |
| Reduce mesh resolution from 2500 to 500 vertices (ablation) | GCGP · MPI Faust | error rate: 3.33% -> 23.33% | Hurt |
Closest papers by meaning
- Gaussian-Induced Convolution for Graphs (AAAI 2019 · similarity 0.90)
- Graph Neural Network-Inspired Kernels for Gaussian Processes in Semi-Supervised Learning (ICLR 2023 · similarity 0.89)
- Bayesian Graph Convolutional Neural Networks for Semi-Supervised Classification (AAAI 2019 · similarity 0.89)
- Bayesian Semi-supervised Learning with Graph Gaussian Processes (NeurIPS 2018 · similarity 0.89)
- Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization (ICML 2021 · similarity 0.88)
- CKGConv: General Graph Convolution with Continuous Kernels (ICML 2024 · similarity 0.88)
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.