Research map / Federated learning and privacy / Personalized and heterogeneous FL
pFL-Bench: A Comprehensive Benchmark for Personalized Federated Learning
Daoyuan Chen, Dawei Gao, Weirui Kuang, Yaliang Li, Bolin Ding
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 local fine-tuning (FT) to FedBN (ablation) | FedBN · FEMNIST s=0.2 un-participated clients | accuracy: 7.86 -> 82.87 | Improved |
| add fine-tuning to FedAvg (ablation) | FedAvg · FEMNIST s=0.2 participated clients | accuracy: 83.97 -> 86.44 | Improved |
| add fine-tuning to FedAvg (ablation) | FedAvg · SST-2 participated clients | accuracy: 74.88 -> 74.14 | No clear effect |
| add fine-tuning to FedEM (ablation) | FedEM · SST-2 participated clients | accuracy: 75.78 -> 64.86 | Hurt |
| add FedBN to FedEM (ablation) | FedEM · FEMNIST s=0.2 un-participated clients | accuracy: 82.81 -> 12.88 | Hurt |
| add FedBN to Ditto (ablation) | Ditto · FEMNIST s=0.2 participated clients | accuracy: 88.39 -> 88.94 | Improved |
| use personalized method Ditto instead of FedAvg | FedAvg · FEMNIST s=0.2 un-participated clients | accuracy: 81.97 -> 2.20 | Hurt |
| use personalized method Ditto instead of FedAvg | FedAvg · FEMNIST s=0.2 participated clients | accuracy: 83.97 -> 88.39 | Improved |
| federated training instead of isolated per-client training | Isolated · SST-2 bottom client accuracy | bottom accuracy: 0.00 -> 44.85 | Improved |
| increase Non-IID degree (Dirichlet alpha from 5 to 0.1) | pFL methods · CIFAR10 | accuracy | Hurt |
Closest papers by meaning
- FedAS: Bridging Inconsistency in Personalized Federated Learning (CVPR 2024 · similarity 0.91)
- Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees (ICML 2025 · similarity 0.89)
- PeFLL: Personalized Federated Learning by Learning to Learn (ICLR 2024 · similarity 0.89)
- pFedClub: Controllable Heterogeneous Model Aggregation for Personalized Federated Learning (NeurIPS 2024 · similarity 0.89)
- $\texttt{pfl-research}$: simulation framework for accelerating research in Private Federated Learning (NeurIPS 2024 · similarity 0.89)
- Flexible Sharpness-Aware Personalized Federated Learning (AAAI 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.
