Research map / Graph learning / GNN architectures
Generalizing Aggregation Functions in GNNs: Building High Capacity and Robust GNNs via Nonlinear Aggregation
Beibei Wang, Bo Jiang, Jin Tang, Bin Luo
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 |
|---|---|---|---|
| replace mean aggregation with L_p-norm aggregation | GCN · Cora | accuracy (%): 81.24 -> 82.57 | Improved |
| replace mean aggregation with polynomial aggregation | GCN · Amazon Computers | accuracy (%): 80.02 -> 82.86 | Improved |
| replace mean aggregation with softmax aggregation | GCN · Cora | accuracy (%): 81.24 -> 83.10 | Improved |
| replace aggregation with L_p-norm aggregation | Masked GCN · Amazon Computers | accuracy (%): 76.06 -> 80.63 | Improved |
| replace aggregation with softmax aggregation | Masked GCN · Amazon Computers | accuracy (%): 76.06 -> 81.50 | Improved |
| replace aggregation with polynomial aggregation | GAT · Amazon Computers | accuracy (%): 79.72 -> 82.60 | Improved |
| replace aggregation with softmax aggregation | CAT · Cora | accuracy (%): 82.45 -> 83.34 | Improved |
| replace aggregation with polynomial aggregation | CAT · Citeseer | accuracy (%): 72.46 -> 71.78 | Hurt |
| replace aggregation with polynomial aggregation | CAT · Cora (clustering) | accuracy (%): 81.98 -> 82.10 | No clear effect |
| replace aggregation with softmax aggregation | GCN · PPI (inductive) | accuracy (%): 97.25 -> 98.70 | Improved |
| replace aggregation with softmax aggregation | Masked GCN · PPI (inductive) | accuracy (%): 97.30 -> 97.95 | Improved |
| replace aggregation with L_p-norm aggregation | GCN · Cora (clustering) | accuracy (%): 80.24 -> 82.40 | Improved |
Closest papers by meaning
- Meta-Aggregator: Learning To Aggregate for 1-Bit Graph Neural Networks (ICCV 2021 · similarity 0.91)
- Sequential Signal Mixing Aggregation for Message Passing Graph Neural Networks (NeurIPS 2024 · similarity 0.89)
- Covered Forest: Fine-grained generalization analysis of graph neural networks (ICML 2025 · similarity 0.88)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.88)
- Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence (ICML 2025 · similarity 0.87)
- Cooperative Graph Neural Networks (ICML 2024 · 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.
