Research map / Graph learning / Graph algorithms and clustering
COPT: Coordinated Optimal Transport on Graphs
Yihe Dong, Will Sawin
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 |
|---|---|---|---|
| use COPT sketches instead of OTC for 4X compression | SVM on sketched graphs · BZR_MD | accuracy (%): 64.3±2.7 -> 68.4±5.0 | Improved |
| use COPT instead of REC at 4X compression | sketching · Enzymes | accuracy (%): 28.9±1.8 -> 33.1±4.2 | Improved |
| use COPT instead of Algebraic at 2X compression | sketching · Enzymes | accuracy (%): 35.1±2.3 -> 32.2±3.3 | Hurt |
| use COPT instead of REC at 2X compression | sketching · Enzymes | accuracy (%): 34.7±2.4 -> 32.2±3.3 | Hurt |
| use COPT instead of OTC at 4X compression | sketching · MSRC_9 | accuracy (%): 84.8±6.7 -> 81.2±4.8 | Hurt |
| use COPT instead of OTC at 2X compression | sketching · Proteins | accuracy (%): 72.8±.8 -> 74.0±1.3 | Improved |
| use COPT sketches instead of spectral projections for retrieval | l1/l2 nearest neighbor · synthetic graph retrieval | accuracy (%): 82.8 ± .6 -> 97.8 ± 1.1 | Improved |
| replace GW on full graphs with COPT sketches | retrieval · synthetic graph retrieval | accuracy (%): 99.8 ± .3 -> 97.8 ± 1.1 | Hurt |
| replace GW on full graphs with COPT sketches | retrieval · synthetic graph retrieval | time: 3.69 ± .07 s -> 1.81 ± .07 ms | Improved |
| filter with COPT instead of spectral projections before GW | COPT+GW pipeline · synthetic graph retrieval, top 3 | accuracy (%): 89.4 ± 2.4 -> 98.7 ± 0.3 | Improved |
| use COPT instead of OTC for 7-node reduction | retrieval · BZR_MD | accuracy (%): 52.2 ± 5.5 -> 57.2 ± 4.9 | Improved |
Closest papers by meaning
- GOT: An Optimal Transport framework for Graph comparison (NeurIPS 2019 · similarity 0.91)
- Hypergraph Learning for Unsupervised Graph Alignment via Optimal Transport (AAAI 2025 · similarity 0.86)
- An Efficient Orlicz-Sobolev Approach for Transporting Unbalanced Measures on a Graph (NeurIPS 2025 · similarity 0.86)
- Generalized Sobolev Transport for Probability Measures on a Graph (ICML 2024 · similarity 0.85)
- Unsupervised Learning for Optimal Transport plan prediction between unbalanced graphs (NeurIPS 2025 · similarity 0.85)
- Learning Ultrametric Trees for Optimal Transport Regression (AAAI 2024 · 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.
