Research map / Graph learning / GNN architectures
Let’s Agree to Degree: Comparing Graph Convolutional Networks in the Message-Passing Framework
Floris Geerts, Filip Mazowiecki, Guillermo Perez
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 |
|---|---|---|---|
| allow message functions to use vertex degrees | MPNNs · labelled graphs | distinguishing power: one step ahead | Improved |
| restrict message functions to depend only on vertex labels, not degrees | anonymous MPNNs · labelled graphs | distinguishing power | No clear effect |
| use a plain Kipf-Welling GCN instead of WL refinement | Kipf-Welling GCN · labelled graphs | distinguishing power | Hurt |
| add a learnable trade-off parameter between a vertex's own features and its neighbours' features | GCNs · labelled graphs | distinguishing power | Improved |
| use ReLU GNNs with a learnable self-feature trade-off parameter | GNNs · fixed input graph | WL simulation rounds: one layer in each step | Improved |
| use dGNN_1 or dGNN_3 rather than a general degree-aware MPNN | dGNN_1 and dGNN_3 · labelled graphs | distinguishing power | Hurt |
| use dGNN_2, dGNN_4, dGNN_5, or dGNN_6 | dGNN architectures · labelled graphs | distinguishing power | Mixed |
Closest papers by meaning
- Analyzing the Expressive Power of Graph Neural Networks in a Spectral Perspective (ICLR 2021 · similarity 0.90)
- Breaking the Limits of Message Passing Graph Neural Networks (ICML 2021 · similarity 0.90)
- Graph Neural Networks with Local Graph Parameters (NeurIPS 2021 · similarity 0.90)
- Fine-grained Expressivity of Graph Neural Networks (NeurIPS 2023 · similarity 0.90)
- A New Perspective on "How Graph Neural Networks Go Beyond Weisfeiler-Lehman?" (ICLR 2022 · similarity 0.90)
- How hard is to distinguish graphs with graph neural networks? (NeurIPS 2020 · 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.
