Research map / Graph learning / Graph algorithms and clustering
Solving graph compression via optimal transport
Vikas Garg, Tommi Jaakkola
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 |
|---|---|---|---|
| OTC instead of best baseline, MSRC-21C | graph compression · MSRC-21C | test accuracy: .485 -> .548 | Improved |
| OTC instead of REC, MSRC-9 at 20% train | graph compression · MSRC-9 | test accuracy: .738 -> .784 | Improved |
| OTC instead of Alg. Dist. on DHFR at 70% train | graph compression · DHFR | test accuracy: .817 -> .814 | No clear effect |
| OTC instead of REC on BZR-MD at 40% train | graph compression · BZR-MD | test accuracy: .563 -> .547 | Hurt |
| OTC instead of Heavy on Mutagenicity at 20% train | graph compression · Mutagenicity | test accuracy: .718 -> .749 | Improved |
| OTC instead of baselines on compression time | graph compression · standard graph datasets | compression time | Improved |
| label-aware edge costs (c=0.01 same label, 0.02 different) | OTC · synthetic tree and real graphs | qualitative compression | Improved |
Closest papers by meaning
- Partition and Code: learning how to compress graphs (NeurIPS 2021 · similarity 0.89)
- COPT: Coordinated Optimal Transport on Graphs (NeurIPS 2020 · similarity 0.87)
- Learning Graph Representation via Graph Entropy Maximization (ICML 2024 · similarity 0.85)
- An Efficient Orlicz-Sobolev Approach for Transporting Unbalanced Measures on a Graph (NeurIPS 2025 · similarity 0.84)
- Efficient Streaming Algorithms for Graphlet Sampling (NeurIPS 2024 · similarity 0.84)
- Submodular framework for structured-sparse optimal transport (ICML 2024 · 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.
