Research map / Graph learning / Graph algorithms and clustering
Generative Graph Dictionary Learning
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 |
|---|---|---|---|
| Nonlinear RBF-kernel generative FGW mixture (FRAME) vs linear reconstructive GDL | graph clustering · ENZYMES, IMDB-M, PTC-MR | Rand Index: up to 0.5% over best competitor | Helped |
| Adding semi-supervised labels (ss-FRAME) to FRAME (ablation) | graph classification · ENZYMES, IMDB-M, PTC-MR | accuracy: ENZYMES 30.2 to 46.3; IMDB-M 40.6 to 51.9 | Helped |
| Unsupervised FRAME vs other GDL methods for graph classification | graph classification · ENZYMES, IMDB-M, PTC-MR | accuracy | No clear effect |
| FRAME node-level embedding vs six node embedding baselines | node clustering · AIDS, ENZYMES, PROTEINS, PTC-MR | Rand Index: up to 2.5% over best competitor | Helped |
| Larger RBF length-scale sigma (ablation) | FRAME graph embeddings · synthetic SBM graphs | embedding sparsity and block structure | Mixed |
| Nonlinear embedding (GWF, FRAME) vs linear GDL for reconstruction (ablation) | graph reconstruction · DBLP, IMDB-M and others | FGW reconstruction error | Helped |
| EM-based FRAME optimization vs gradient-based GDL | running time · graph reconstruction benchmarks | running time | Helped |
| Pearson correlation between graph-space FGW distances and embedding-space distances | FRAME embeddings · DBLP, IMDB-M | Pearson coefficient | No clear effect |
Closest papers by meaning
- Online Graph Dictionary Learning (ICML 2021 · similarity 0.91)
- Robust Graph Dictionary Learning (ICLR 2023 · similarity 0.87)
- Semi-relaxed Gromov-Wasserstein divergence and applications on graphs (ICLR 2022 · similarity 0.87)
- THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplings (AAAI 2025 · similarity 0.83)
- StructComp: Substituting propagation with Structural Compression in Training Graph Contrastive Learning (ICLR 2024 · similarity 0.83)
- S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning (ICML 2024 · similarity 0.83)
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.