Research map / Graph learning / GNN architectures
Neural Dispersion on Graphs
Ryien Hosseini, Pouya Mahdi Gholami, Filippo Simini, Venkatram Vishwanath, Rebecca Willett, Henry Hoffmann
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 generator ensemble instead of single generator (ablation) | NGD · Graph dispersion N=16 | GCD diversity: 6.100 -> 7.621 | Improved |
| use directional random-projection metrics instead of a single scalar projection (ablation) | NGD · Graph dispersion N=16 | GCD diversity: 6.242 -> 7.621 | Improved |
| use generator ensemble instead of single generator (ablation) | NGD · Graph dispersion N=128 | Portrait diversity: 0.929 -> 0.969 | Improved |
| NGD instead of ER-mix parametric ensemble | NGD · Graph dispersion N=128 | Heat diversity: 0.068 -> 5.546 | Improved |
| NGD instead of combinatorial search at small N | NGD · Graph dispersion N=16 | GCD diversity: 7.718 (Greedy->Genetic->Local) vs 7.621 | Mixed |
| NGD instead of combinatorial search at N=128 | NGD · Graph dispersion N=128 | Wave diversity: 10.974 (Greedy->Genetic) vs 10.674 | Hurt |
| scale N and k with NGD | NGD · Large N up to 1024, k up to 8192 | diversity | Improved |
| evaluate GIN on NGD graphs instead of generator-mix graphs | GIN regressor · Spectral radius, N=256 | MAE: 3.2956 -> 10.5067 | Hurt |
| evaluate greedy coloring on NGD graphs instead of generator-mix | greedy coloring · Approximation ratio, largest-first | mean ratio: 0.25 -> 0.48 | Hurt |
Closest papers by meaning
- Challenges of Generating Structurally Diverse Graphs (NeurIPS 2024 · similarity 0.91)
- Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuning (ICLR 2025 · similarity 0.87)
- Towards Addressing Frontiers in Graph Generation (AAAI 2025 · similarity 0.86)
- On the Role of Edge Dependency in Graph Generative Models (ICML 2024 · similarity 0.86)
- HiGen: Hierarchical Graph Generative Networks (ICLR 2024 · similarity 0.86)
- How Graph Neural Networks Learn: Lessons from Training Dynamics (ICML 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.
