Research map / Graph learning / GNN architectures
Equivariant Neural Operator Learning with Graphon Convolution
Chaoran Cheng, Jian Peng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 residual operator layer (vs no-res) (ablation) | InfGCN · QM9 rotated | NMAE (%): 6.14 -> 4.73 | Improved |
| add residual operator layer (vs no-res) (ablation) | InfGCN · QM9 unrotated | NMAE (%): 3.72 -> 0.93 | Improved |
| use channel-wise tensor product instead of fully-connected (ablation) | InfGCN · QM9 rotated | NMAE (%): 4.95 -> 4.73 | Improved |
| use channel-wise tensor product instead of fully-connected (ablation) | InfGCN · QM9 unrotated | NMAE (%): 1.36 -> 0.93 | Improved |
| channel-wise instead of fully-connected: parameter count (ablation) | InfGCN · QM9 | NMAE (%): 17.42 -> 1.20 | Improved |
| raise max spherical degree from L=0 to L=7 (ablation) | InfGCN · QM9 rotated | NMAE (%): 12.62 -> 4.73 | Improved |
| raise max spherical degree from L=0 to L=7 (ablation) | InfGCN · QM9 unrotated | NMAE (%): 12.05 -> 0.93 | Improved |
| raise max spherical degree from L=4 to L=7 (ablation) | InfGCN · QM9 rotated | NMAE (%): 4.77 -> 4.73 | No clear effect |
| raise max spherical degree from L=4 to L=7 (ablation) | InfGCN · QM9 unrotated | NMAE (%): 1.08 -> 0.93 | No clear effect |
| raise max spherical degree from L=2 to L=3 (ablation) | InfGCN · QM9 rotated | NMAE (%): 6.95 -> 4.86 | Improved |
| raise max spherical degree from L=1 to L=2 (ablation) | InfGCN · QM9 unrotated | NMAE (%): 8.07 -> 4.65 | Improved |
| InfGCN vs second-best DeepDFT2 (graphon conv + residual) | InfGCN · Cubic | NMAE (%): 10.37 -> 8.98 | Improved |
Closest papers by meaning
- Equivariant Graph Neural Operator for Modeling 3D Dynamics (ICML 2024 · similarity 0.87)
- E2Former: An Efficient and Equivariant Transformer with Linear-Scaling Tensor Products (NeurIPS 2025 · similarity 0.87)
- Equivariant Machine Learning on Graphs with Nonlinear Spectral Filters (NeurIPS 2024 · similarity 0.86)
- Equivariant Matrix Function Neural Networks (ICLR 2024 · similarity 0.86)
- On the Generalization of Equivariant Graph Neural Networks (ICML 2024 · similarity 0.85)
- SEGNO: Generalizing Equivariant Graph Neural Networks with Physical Inductive Biases (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.
