Research map / Graph learning / GNN architectures
MGNNI: Multiscale Graph Neural Networks with Implicit Layers
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 multiscale {1,2} instead of single scale {1} (ablation) | MGNNI · Chameleon | accuracy (%): 61.46 -> 63.93 | Helped |
| use multiscale {1,2,3} instead of single scale {1} (ablation) | MGNNI · Texas | accuracy (%): 81.35 -> 84.86 | Helped |
| use multiscale {1,2} instead of single scale {1} (ablation) | MGNNI · PPI | Micro-F1 (%): 98.35 -> 98.67 | Helped |
| use only 3-hop propagation instead of 1-hop (ablation) | MGNNI · PPI | Micro-F1 (%): 98.35 -> 88.56 | Hurt |
| use only 3-hop propagation instead of 1-hop (ablation) | MGNNI · Texas | accuracy (%): 81.35 -> 83.78 | Helped |
| add third scale to {1,2} (ablation) | MGNNI · Chameleon | accuracy (%): 63.93 -> 63.75 | No clear effect |
| use MGNNI instead of EIGNN | implicit GNN · Squirrel | accuracy (%): 46.37 -> 54.50 | Helped |
| use MGNNI instead of EIGNN | implicit GNN · Chameleon | accuracy (%): 62.92 -> 63.93 | Helped |
| use MGNNI instead of EIGNN | implicit GNN · Wisconsin | accuracy (%): 86.86 -> 86.67 | No clear effect |
| use MGNNI instead of IGNN | implicit GNN · PPI | Micro-F1 (%): 97.6 -> 98.7 | Helped |
| use MGNNI instead of EIGNN | implicit GNN · PPI | Micro-F1 (%): 98.0 -> 98.7 | Helped |
| use MGNNI instead of GIN | graph classification · NCI1 | accuracy (%): 82.7 -> 78.9 | Hurt |
Closest papers by meaning
- Scalable and Effective Implicit Graph Neural Networks on Large Graphs (ICLR 2024 · similarity 0.89)
- Demystifying Structural Disparity in Graph Neural Networks: Can One Size Fit All? (NeurIPS 2023 · similarity 0.89)
- Graph Neural Networks Inspired by Classical Iterative Algorithms (ICML 2021 · similarity 0.89)
- Implicit Graph Neural Networks (NeurIPS 2020 · similarity 0.89)
- T2-GNN: Graph Neural Networks for Graphs with Incomplete Features and Structure via Teacher-Student Distillation (AAAI 2023 · similarity 0.88)
- GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural Network for Heterophilous Graphs (AAAI 2025 · 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.