Research map / Graph learning / GNN architectures
Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
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 multi-scale concatenation in truncated Krylov layers instead of standard shallow GCN propagation | truncated Krylov · Cora, CiteSeer, and PubMed, no-validation splits | accuracy | Helped |
| Use dense Snowball concatenation with identity activations rather than Tanh activations | Snowball · Cora, CiteSeer, and PubMed, no-validation splits | accuracy | Mixed |
| Replace ReLU with Tanh | deep GCN · 100-layer synthetic graph experiment | hidden-feature rank | Helped |
| Add concatenation of earlier hidden features at each layer | Snowball GCN · Cora, CiteSeer, and PubMed node classification | accuracy | Helped |
Closest papers by meaning
- Simplifying Graph Convolutional Networks (ICML 2019 · similarity 0.92)
- Simple Spectral Graph Convolution (ICLR 2021 · similarity 0.92)
- Simple and Deep Graph Convolutional Networks (ICML 2020 · similarity 0.91)
- All in a Row: Compressed Convolution Networks for Graphs (ICML 2023 · similarity 0.91)
- Deeper Insights Into Graph Convolutional Networks for Semi-Supervised Learning (AAAI 2018 · similarity 0.91)
- Generalizing CNNs to graphs with learnable neighborhood quantization (NeurIPS 2024 · similarity 0.89)
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.