Research map / Graph learning / GNN architectures
Graph Filtration Kernels
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 WL filtration with degree-based edge weights instead of ordinary WL | WL filtration kernel · EGO-2 | accuracy | Helped |
| Use WL filtration with degree-based edge weights instead of ordinary WL | WL filtration kernel · EGO-3 | accuracy | Helped |
| Use WL filtration with degree-based edge weights instead of ordinary WL | WL filtration kernel · EGO-4 | accuracy: ~10% accuracy increase | Helped |
| Use WL filtration with degree-based edge weights | WL filtration kernel · EGO-1 | accuracy | Mixed |
| Use WL filtration kernels on the molecular and IMDB benchmarks | WL filtration kernel · DHFR, NCI1, PTC-MR, IMDB-BINARY | accuracy | No clear effect |
| Increase filtration length from k=1 to k=2 (ablation) | WL filtration kernel with degree-based edge weights · EGO datasets | accuracy | Helped |
| Vary filtration length beyond k=2 or k=3 (ablation) | WL filtration kernel with degree-based edge weights · EGO datasets | accuracy | Hurt |
| Increase random-walk length from λ=1 to larger values (ablation) | WL filtration kernel with random-walk edge weights · CSL graphs | accuracy | Helped |
| Use λ=7 random-walk edge weights (ablation) | WL filtration kernel with random-walk edge weights · CSL graphs | graph distinguishability | Helped |
Closest papers by meaning
- Rethinking Kernel Methods for Node Representation Learning on Graphs (NeurIPS 2019 · similarity 0.91)
- Convolutional Kernel Networks for Graph-Structured Data (ICML 2020 · similarity 0.91)
- Wasserstein Weisfeiler-Lehman Graph Kernels (NeurIPS 2019 · similarity 0.90)
- Exploring Consistency in Graph Representations: from Graph Kernels to Graph Neural Networks (NeurIPS 2024 · similarity 0.90)
- Generalizing Weisfeiler-Lehman Kernels to Subgraphs (ICLR 2025 · similarity 0.89)
- DHAKR: Learning Deep Hierarchical Attention-Based Kernelized Representations for Graph Classification (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.