Research map / Graph learning / Graph algorithms and clustering
The Statistical Performance of Collaborative Inference
Gérard Biau, Kevin Bleakley, Benoît Cadre
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 3-regular Ramanujan graph instead of sparser non-Ramanujan graphs | collaborative inference · simulation, N up to 1000 | performance ratio tau_t | Improved |
| use 5-regular Ramanujan graph instead of 3-regular | collaborative inference · simulation, N=1000 | performance ratio tau_t | Improved |
| increase degree d of Ramanujan graph (ablation) | collaborative inference · N=200 | statistical coefficient S(A_d) | Improved |
| increase penalty beta on connections (ablation) | penalized objective S+beta*C · N=200, beta in {1/2,1,2,4} | optimal degree d* | Hurt |
| increase number of nodes N beyond 710 for fixed T=100 million (ablation) | 3-regular Ramanujan network · streaming, tau >= 0.99 | cost (degree d* needed): d* 3 -> 4, 5 | Hurt |
Closest papers by meaning
- Joint Inference of Multiple Graphs from Matrix Polynomials (JMLR 2022 · similarity 0.87)
- Scalable Decentralized Algorithms for Online Personalized Mean Estimation (AAAI 2025 · similarity 0.86)
- Mutual information for symmetric rank-one matrix estimation: A proof of the replica formula (NeurIPS 2016 · similarity 0.86)
- Fundamental Limits of Distributed Covariance Matrix Estimation Under Communication Constraints (ICML 2024 · similarity 0.86)
- Exact Community Recovery under Side Information: Optimality of Spectral Algorithms (ICLR 2025 · similarity 0.85)
- Sample Complexity of Correlation Detection in the Gaussian Wigner Model (ICML 2025 · 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.
