Research map / Graph learning / Graph algorithms and clustering
GraphOpt: Learning Optimization Models of Graph Formation
Rakshit Trivedi, Jiachen Yang, Hongyuan Zha
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 GraphOpt-Embed rather than SEAL | Cora-ML link prediction · Cora-ML | AUC: 97.21 -> 96.21 | Hurt |
| Use GraphOpt-Embed rather than SEAL | Political Blogs link prediction · Political Blogs | AUC: 95.32 -> 95.50 | Improved |
| Use GraphOpt-Embed rather than SEAL | E. Coli link prediction · E. Coli | AUC: 97.12 -> 97.20 | Improved |
| Use GraphOpt-Embed rather than the strongest baseline | WN18RR link prediction · WN18RR | MRR: 47.2 -> 47.3 | Improved |
| Transfer the learned Cora-ML reward and train a new policy, rather than directly deploy the Cora-ML policy | CiteSeer graph construction · CiteSeer | Triangle Count: 2234 ± 58.96 -> 2847.66 ± 57.13 | Improved |
| Transfer the learned Cora-ML reward and train a new policy, rather than directly deploy the Cora-ML policy | CiteSeer graph construction · CiteSeer | Clustering Coeff.: 0.084 ± 0.004 -> 0.098 ± 0.0010 | Improved |
| Transfer the learned Cora-ML reward and train a new policy, rather than directly deploy the Cora-ML policy | CiteSeer graph construction · CiteSeer | Max Degree: 70.66 ± 2.081 -> 80.66 ± 1.527 | Improved |
| Learn and optimize a latent reward instead of directly transferring the trained policy | CiteSeer graph construction · CiteSeer | Triangle Count: 2234 ± 58.96 -> 2847.66 ± 57.13 | Improved |
| Use GraphOpt instead of NetGAN for graph synthesis | Barabasi Albert graph construction · Barabasi Albert | Triangle Count deviation (%): 31.68 ± 6.28 -> 6.28 ± 4.05 | Improved |
| Use GraphOpt instead of NetGAN for graph synthesis | Political Blogs graph construction · Political Blogs | Clustering Coeff. deviation (%): 37.55 ± 7.2 -> 20.34 ± 9.1 | Improved |
| Use a learned embedding for prediction rather than the GraphOpt construction policy | Cora-ML link prediction · Cora-ML | AUC: 93.50 -> 96.21 | Improved |
| Use a learned embedding for prediction rather than the GraphOpt construction policy | Political Blogs link prediction · Political Blogs | AUC: 92.21 -> 95.50 | Improved |
Closest papers by meaning
- Graph Reinforcement Learning for Network Control via Bi-Level Optimization (ICML 2023 · similarity 0.87)
- Beyond Node Sequences: Relational Diffusion for Unified Graph Learning (NeurIPS 2026 · similarity 0.86)
- Transfer Learning on Edge Connecting Probability Estimation Under Graphon Model (NeurIPS 2025 · similarity 0.85)
- Graph Cascades: Contagion-Based Mesoscopic Rewiring for Structure-Aware Graph Machine Learning (NeurIPS 2026 · similarity 0.85)
- When Edge Independence Fails: Joint Graph Diffusion with Latent Sociability Priors (NeurIPS 2026 · similarity 0.85)
- Learning Latent Graph Structures and their Uncertainty (ICML 2025 · 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.
