Research map / Graph learning / Graph algorithms and clustering
Generalizing Downsampling from Regular Data to Graphs
Davide Bacciu, Alessio Conte, Francesco Landolfi
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 |
|---|---|---|---|
| Use k-MIS max-pooling instead of EdgePool | Graph classification model · REDDIT-B | accuracy: 85.98 ± 1.57 -> 87.57 ± 1.96 | Improved |
| Use k-MIS strided reduction instead of EdgePool | Graph classification model · REDDIT-5K | accuracy: 52.44 ± 1.11 -> 54.30 ± 0.53 | Improved |
| Use k-MIS mean-pooling instead of the no-pooling baseline | Graph classification model · GITHUB | accuracy: 69.89 ± 0.28 -> 68.60 ± 0.67 | Hurt |
| Use k-MIS pooling instead of the no-pooling baseline | Graph classification model · DD | accuracy: 75.51 ± 1.07 -> 76.91 ± 1.06 | Mixed |
| Use k-MIS pooling instead of the no-pooling baseline | Graph classification model · REDDIT-12K | accuracy: 45.04 ± 6.63 -> 47.51 ± 0.99 | Improved |
Closest papers by meaning
- Taxonomy of reduction matrices for Graph Coarsening (NeurIPS 2025 · similarity 0.87)
- Geometry-Aware Edge Pooling for Graph Neural Networks (NeurIPS 2025 · similarity 0.86)
- UGC: Universal Graph Coarsening (NeurIPS 2024 · similarity 0.86)
- Graph Parsing Networks (ICLR 2024 · similarity 0.86)
- SkipPool: Improved Sparse Hierarchical Graph Pooling with Differentiable Exploration (AAAI 2025 · similarity 0.85)
- Graph Sparsification via Mixture of Graphs (ICLR 2025 · 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.
