Research map / Optimization / Nonconvex and smooth optimization
Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods
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 |
|---|---|---|---|
| GFN instead of untuned Supervised PageRank | Supervised PageRank · Q1_2 | loss: .00357 -> .00274 | Helped |
| GBN instead of GBP 50 steps | GBP · Q1_2 | loss: .00282 -> .00279 | Helped |
| GFN instead of GBP 50 steps | GBP · Q2_2 | loss: .00307 -> .00297 | Helped |
| GBN instead of GBP 50 steps | GBP · Q3_2 | loss: .00295 -> .00295 | No clear effect |
| GBN instead of GBP step size 50 | GBP · Q2_2 | upper-level steps: 31 -> 12 | Helped |
| GFN instead of GBN | GBN · Q3_2 | upper-level steps: 12 -> 106 | Hurt |
| increase GBP step size from 50 to 500 | GBP · Q1_2 | loss: .00282 -> .00283 | No clear effect |
| increase GBP step size from 50 to 500 | GBP · Q1_2 | upper-level steps: 16 -> 2 | Helped |
| vary GBN initial Lipschitz constant L0 from 1e-4 to 1 (ablation) | GBN · all test sets | loss: differences are less than 10−7 | No clear effect |
| tighten accuracy epsilon below 1e-6 (ablation) | GBN and GFN · Q1_2-Q3_2 | loss | No clear effect |
Closest papers by meaning
- Distributed PageRank Computation: An Improved Theoretical Study (AAAI 2019 · similarity 0.86)
- Accelerated Evolving Set Processes for Local PageRank Computation (NeurIPS 2025 · similarity 0.85)
- Complexity of Classical Acceleration for $\ell_1$-Regularized PageRank (NeurIPS 2026 · similarity 0.85)
- Fast Algorithms for Hypergraph PageRank with Applications to Semi-Supervised Learning (ICML 2024 · similarity 0.84)
- PageRank Bandits for Link Prediction (NeurIPS 2024 · similarity 0.84)
- Optimizing Recall in Deep Graph Hashing Framework for Item Retrieval (Student Abstract) (AAAI 2024 · similarity 0.81)
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.