Research map / Federated learning and privacy / Personalized and heterogeneous FL
Generalizable Heterogeneity-aware Federated Feature and Basic-matrix Consistency Learning
Xuan Lai, Luying Zhong, Tianying Lu, Junjie Zhang, Zhiqin Huang, Zheyi Chen
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 FIA feature alignment on public data to local training (ablation) | local baseline · Digits intra AVG | accuracy: 84.59 -> 90.89 | Improved |
| add FIA alignment (ablation) | local baseline · Digits inter AVG | accuracy: 26.00 -> 30.16 | Improved |
| add FBM basis-matrix consistency on top of FIA (ablation) | FIA · Digits inter AVG | accuracy: 30.16 -> 31.77 | Improved |
| add FBM on top of FIA (ablation) | FIA · Digits intra AVG | accuracy: 90.89 -> 90.78 | No clear effect |
| add CND distillation (ablation) | FIA+FBM · Digits | intra/inter accuracy | Mixed |
| add FRR feature relationship regularization (ablation) | FIA+FBM · Digits inter | accuracy | Improved |
| add CND and FRR together (full FBCL) (ablation) | FIA+FBM · Digits | intra AVG / inter AVG: 90.78 -> 93.10 intra; 31.77 -> 34.84 inter | Improved |
| FBCL vs FCCL+ | FCCL+ · Office-Caltech inter-domain AVG | accuracy: 26.00 -> 27.51 | Improved |
| FBCL vs FCCL on Office-Caltech intra AVG | FCCL · Office-Caltech intra AVG | accuracy: 67.77 -> 67.98 | No clear effect |
| FBCL vs FedPSA under model homogeneity | FedPSA · Office-Caltech inter AVG (ResNet-18) | accuracy: 68.80 -> 68.99 | No clear effect |
| collaborative learning vs local baseline | local baseline · Office-Caltech intra AVG | accuracy: 66.08 -> 62.64 (FedMD) | Mixed |
Closest papers by meaning
- Learn From Others and Be Yourself in Heterogeneous Federated Learning (CVPR 2022 · similarity 0.92)
- 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)
- Accurate Forgetting for Heterogeneous Federated Continual Learning (ICLR 2024 · similarity 0.87)
- Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchor (CVPR 2025 · similarity 0.86)
- FedSPA: Generalizable Federated Graph Learning under Homophily Heterogeneity (CVPR 2025 · similarity 0.86)
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.
