Research map / Graph learning / Graph self-supervised learning
Joint Feature and Differentiable $ k $-NN Graph Learning using Dirichlet Energy
Lei Xu, Lei Chen, Rong Wang, Feiping Nie, Xuelong Li
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 |
|---|---|---|---|
| remove differentiable graph learning (DGL) (ablation) | Ours · Madelon | classification ACC: 0.90 -> 0.51 | Hurt |
| remove differentiable graph learning (DGL) (ablation) | Ours · GLIOMA | classification ACC: 0.81 -> 0.66 | Hurt |
| remove differentiable graph learning (DGL) (ablation) | Ours · PCMAC | classification ACC: 0.79 -> 0.72 | Hurt |
| remove DGL (ablation) | Ours · Jaffe | classification ACC: 1.00 -> 0.97 | No clear effect |
| use joint learning instead of Heat kernel graph (ablation) | spectral clustering · PIX10 | clustering ACC: 0.41 -> 0.77 | Improved |
| use joint learning instead of Heat kernel graph (ablation) | spectral clustering · PCMAC | clustering ACC: 0.51 -> 0.51 | No clear effect |
| DGL only (no feature selection) versus full method (ablation) | spectral clustering · Jaffe | clustering ACC: 0.80 -> 0.57 | Hurt |
| DGL only versus Heat kernel graph (ablation) | spectral clustering · PIX10 | clustering ACC: 0.41 -> 0.76 | Improved |
| DGL only versus Heat kernel graph (ablation) | spectral clustering · Madelon | clustering ACC: 0.54 -> 0.50 | Hurt |
| remove unique feature selector (UFS) from network (ablation) | Ours · toy Blobs/Moons/Circles | selected features | Hurt |
| replace Heat-kernel Dirichlet energy FS (DUFS) with Ours | DUFS · Yale | classification ACC: 0.72 -> 0.81 | Improved |
| Ours versus DUFS in clustering | DUFS · SMK | clustering ACC: 0.58 -> 0.64 | Improved |
Closest papers by meaning
- Adaptive Node Feature Selection for Graph Neural Networks (ICML 2026 · similarity 0.87)
- NAFS: A Simple yet Tough-to-beat Baseline for Graph Representation Learning (ICML 2022 · similarity 0.86)
- How Graph Neural Networks Learn: Lessons from Training Dynamics (ICML 2024 · similarity 0.86)
- Joint Graph Rewiring and Feature Denoising via Spectral Resonance (ICLR 2025 · similarity 0.85)
- Improving Graph Neural Networks by Learning Continuous Edge Directions (ICLR 2025 · similarity 0.85)
- Asymmetric Learning for Spectral Graph Neural Networks (AAAI 2025 · similarity 0.84)
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.
