Research map / Federated learning and privacy / Personalized and heterogeneous FL
Heterogeneity-aware Clustered Distributed Learning for Multi-source Data Analysis
Yuanxing Chen, Qingzhao Zhang, Shuangge Ma, Kuangnan Fang
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 ICR fused clustering penalty instead of Local per-client fitting | ICR · Example 1 (p=100, K=64) | model size (true 16): 1056.990 -> 16.000 | Improved |
| use ICR instead of SK(har) two-step clustering | ICR · Example 1 (p=100, K=64) | ARI: 0.462 -> 1.000 | Improved |
| use ICR instead of SK(gap) two-step clustering | ICR · Example 1 (p=100, K=64) | FPR: 0.964 -> 0.000 | Improved |
| use ICR instead of SK(gap) for clustering accuracy | ICR · Example 1 (p=100, K=64) | ARI: 1.000 -> 1.000 | No clear effect |
| use ICR instead of SHIR full pooling | ICR · Example 1 (p=100, K=64) | model size (true 16): 512.670 -> 16.000 | Improved |
| increase number of clients K from 16 to 32 | ICR · Example 1 (p=500) | TPR: 0.921 -> 0.999 | Improved |
| increase number of clients K from 16 to 64 | ICR · Example 1 (p=100) | FPR: 0.000 -> 0.000 | No clear effect |
| increase dimension p from 100 to 500 | ICR · Example 1 (K=16) | TPR: 0.990 -> 0.921 | Hurt |
| use Local estimation instead of SHIR integration | Local · Example 1 (p=100, K=64) | TPR: 1.000 -> 0.891 | Hurt |
| use ICR instead of SK(har) clustering | ICR · Example 1 (p=500, K=16) | ARI: 0.570 -> 1.000 | Improved |
Closest papers by meaning
- Clustered Federated Learning via Gradient-based Partitioning (ICML 2024 · similarity 0.89)
- Enhancing Clustered Federated Learning: Integration of Strategies and Improved Methodologies (ICLR 2025 · similarity 0.88)
- An Aggregation-Free Federated Learning for Tackling Data Heterogeneity (CVPR 2024 · similarity 0.87)
- FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning (NeurIPS 2025 · similarity 0.87)
- Heterogeneity-Guided Client Sampling: Towards Fast and Efficient Non-IID Federated Learning (NeurIPS 2024 · similarity 0.87)
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 2025 · similarity 0.87)
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.
