Research map / Federated learning and privacy / Personalized and heterogeneous FL
PFLlib: A Beginner-Friendly and Comprehensive Personalized Federated Learning Library and Benchmark
Jianqing Zhang, Yang Liu, Yang Hua, Hao Wang, Tao Song, Zhengui Xue, Ruhui Ma, Jian 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 |
|---|---|---|---|
| use personalized FL (FedCP) instead of FedAvg | FedAvg · Pathological label skew Cifar100 | test accuracy (%): 25.98 -> 71.80 | Improved |
| use FedDBE instead of FedAvg | FedAvg · Practical label skew Cifar100 | test accuracy (%): 31.89 -> 64.39 | Improved |
| use GPFL instead of FedAvg | FedAvg · Practical label skew AG News | test accuracy (%): 87.12 -> 97.97 | Improved |
| use FedProx instead of FedAvg | FedAvg · Pathological label skew FMNIST | test accuracy (%): 80.41 -> 78.08 | Hurt |
| use FedGen instead of FedAvg | FedAvg · Pathological label skew Cifar100 | test accuracy (%): 25.98 -> 20.80 | Hurt |
| use FedProx instead of FedAvg | FedAvg · Practical label skew Cifar100 | test accuracy (%): 31.89 -> 31.99 | No clear effect |
| use ResNet-18 instead of 4-layer CNN for FedAvg | FedAvg · Practical label skew Tiny-ImageNet | test accuracy (%): 19.46 -> 19.45 | No clear effect |
| use ResNet-18 instead of 4-layer CNN for FedCP | FedCP · Practical label skew Tiny-ImageNet | test accuracy (%): 43.49 -> 44.18 | No clear effect |
| use ResNet-18 instead of 4-layer CNN for pFedMe | pFedMe · Practical label skew Tiny-ImageNet | test accuracy (%): 26.93 -> 33.44 | Improved |
| use FedALA instead of Per-FedAvg (meta-learning) | Per-FedAvg · Practical label skew Cifar100 | test accuracy (%): 44.28 -> 55.92 | Improved |
| use FedBABU instead of FedAvg | FedAvg · Practical label skew AG News | test accuracy (%): 87.12 -> 95.86 | Improved |
| use FedAMP instead of FedAvg | FedAvg · Practical label skew AG News | test accuracy (%): 87.12 -> 83.35 | Hurt |
Closest papers by meaning
- $\texttt{pfl-research}$: simulation framework for accelerating research in Private Federated Learning (NeurIPS 2024 · similarity 0.88)
- FedAS: Bridging Inconsistency in Personalized Federated Learning (CVPR 2024 · similarity 0.87)
- PeFLL: Personalized Federated Learning by Learning to Learn (ICLR 2024 · similarity 0.86)
- Adaptive Latent-Space Constraints in Personalized Federated Learning (NeurIPS 2025 · similarity 0.86)
- PFedCS: A Personalized Federated Learning Method for Enhancing Collaboration among Similar Classifiers (AAAI 2025 · similarity 0.86)
- Look Back for More: Harnessing Historical Sequential Updates for Personalized Federated Adapter Tuning (AAAI 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.
