Research map / Federated learning and privacy / Personalized and heterogeneous FL
Generalizable Heterogeneous Federated Cross-Correlation and Instance Similarity Learning
Wenke Huang, Mang Ye, Zekun Shi, Bo Du
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 |
|---|---|---|---|
| add FCCM (cross-correlation matrix) (ablation) | SOLO baseline · Digits inter-domain | average accuracy: 26.68 -> 33.51 | Improved |
| add FISL (instance similarity learning) (ablation) | SOLO baseline · Digits inter-domain | average accuracy: 26.68 -> 37.74 | Improved |
| add FISL to FCCM (ablation) | FCCM · Digits inter-domain | average accuracy: 33.51 -> 40.45 | Improved |
| add FNTD in local updating (ablation) | FCCM+FISL · Digits inter-domain | average accuracy: 40.45 -> 50.49 | Improved |
| add FCCM (ablation) | SOLO baseline · Office31 inter-domain | average accuracy: 27.72 -> 33.52 | Improved |
| add FISL alone (ablation) | SOLO baseline · Office31 inter-domain | average accuracy: 27.72 -> 28.19 | No clear effect |
| add FNTD (ablation) | FCCM+FISL · Office31 inter-domain | average accuracy: 32.15 -> 36.36 | Improved |
| add FCCM (ablation) | SOLO baseline · Digits intra-domain | average accuracy: 73.14 -> 85.69 | Improved |
| add FNTD (ablation) | FCCM+FISL · Digits intra-domain | average accuracy: 90.30 -> 90.57 | Improved |
| add FNTD (ablation) | FCCM+FISL · Office31 intra-domain | average accuracy: 77.50 -> 77.59 | No clear effect |
| use strong instead of weak augmentation on public data for FISL (ablation) | FISL · Digits inter-domain (Cifar-100) | average accuracy: 37.74 -> 37.16 | No clear effect |
| vary lambda in FCCM from 0.002 to 0.1 (ablation) | FCCM · Office31 inter-domain | average accuracy: 33.09, 33.52, 32.89, 32.41, 32.51 | No clear effect |
Closest papers by meaning
- Accurate Forgetting for Heterogeneous Federated Continual Learning (ICLR 2024 · similarity 0.88)
- FedLMT: Tackling System Heterogeneity of Federated Learning via Low-Rank Model Training with Theoretical Guarantees (ICML 2024 · similarity 0.88)
- pFedMxF: Personalized Federated Class-Incremental Learning with Mixture of Frequency Aggregation (CVPR 2025 · similarity 0.87)
- One Arrow, Two Hawks: Sharpness-aware Minimization for Federated Learning via Global Model Trajectory (ICML 2025 · similarity 0.87)
- Federated Orthogonal Training: Mitigating Global Catastrophic Forgetting in Continual Federated Learning (ICLR 2024 · similarity 0.87)
- A Bayesian Approach for Personalized Federated Learning in Heterogeneous Settings (NeurIPS 2024 · 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.
