Research map / Federated learning and privacy / Personalized and heterogeneous FL
PFedEdit: Personalized Federated Learning via Automated Model Editing
Haolin Yuan, William Paul, John Aucott, Philippe Burlina, Yinzhi Cao
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 |
|---|---|---|---|
| replace FedAvg with PFedEdit (automated layer editing) at DH-level 100% | FedAvg · CIFAR-100 DH-level 100% | accuracy: 0.664 -> 0.930 | Improved |
| use PFedEdit instead of FedAvg in i.i.d. setting | FedAvg · CIFAR-10 i.i.d. | accuracy: 0.942 -> 0.955 | Improved |
| use FedPer, FedBN or FedRep personalization instead of FedAvg in i.i.d. setting | FedAvg · CIFAR-10 i.i.d. | accuracy: 0.942 -> 0.890 (FedPer) | Hurt |
| use PFedEdit for faster convergence versus FedAvg | FedAvg · Lyme | rounds to reach accuracy above 0.7: 87 -> 4 rounds | Improved |
| increase edited layer ratio from 1% to 7% (ablation) | PFedEdit · CIFAR-10 | accuracy: 0.933 -> 0.962 | Improved |
| increase edited layer ratio from 7% to 9% (ablation) | PFedEdit · CIFAR-100 | accuracy: 0.890 -> 0.874 | Hurt |
| increase representative subset ratio from 0.01 to 0.08 (ablation) | PFedEdit · CIFAR-10 | accuracy: 0.951 -> 0.954 | No clear effect |
| increase representative subset ratio from 0.08 to 0.1 (ablation) | PFedEdit · CIFAR-10 | accuracy: 0.954 -> 0.962 | Improved |
| assess edits with a prediction list instead of accuracy (ablation) | PFedEdit · CIFAR-10 | accuracy: 0.540 -> 0.962 | Improved |
| assess edits with a prediction list instead of loss (ablation) | PFedEdit · CIFAR-10 | accuracy: 0.905 -> 0.962 | Improved |
| assess edits with a prediction list instead of TE score (ablation) | PFedEdit · CIFAR-100 | accuracy: 0.821 -> 0.890 | Improved |
Closest papers by meaning
- pFedClub: Controllable Heterogeneous Model Aggregation for Personalized Federated Learning (NeurIPS 2024 · similarity 0.91)
- Look Back for More: Harnessing Historical Sequential Updates for Personalized Federated Adapter Tuning (AAAI 2025 · similarity 0.89)
- Personalized Federated Learning via Feature Distribution Adaptation (NeurIPS 2024 · similarity 0.89)
- PFedCS: A Personalized Federated Learning Method for Enhancing Collaboration among Similar Classifiers (AAAI 2025 · similarity 0.89)
- FedAS: Bridging Inconsistency in Personalized Federated Learning (CVPR 2024 · similarity 0.89)
- Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees (ICML 2025 · similarity 0.88)
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.
