Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedReLa: Imbalanced Federated Learning via Re-Labeling
Guangzheng Hu, Patricia Menendez, Feng Liu, Mingming Gong, Guanghui Wang, Liuhua Peng
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 FedReLa one-shot re-labeling | FedLOGE · CIFAR-10-LT IF=100 alpha=0.1 | overall accuracy: 45.08 -> 62.16 | Improved |
| add FedReLa one-shot re-labeling | FedLOGE · CIFAR-100-LT IF=100 alpha=0.1 | overall accuracy: 29.21 -> 38.08 | Improved |
| add FedReLa one-shot re-labeling | FedETF · CIFAR-10-LT IF=50 alpha=0.1 | overall accuracy: 47.96 -> 61.71 | Improved |
| add FedReLa to FedAvg | FedAvg · CIFAR-10 step-wise IR=20 10% minority | minority accuracy: 17.25 -> 51.60 | Improved |
| add FedReLa to FedProx | FedProx · CIFAR-10 step-wise IR=20 10% minority | minority accuracy: 15.60 -> 53.90 | Improved |
| add FedReLa to FedAvg | FedAvg · CIFAR-10 step-wise IR=10 30% minority | majority accuracy: 65.93 -> 62.14 | Hurt |
| add FedReLa to FedAvg | FedAvg · CIFAR-10 step-wise IR=10 10% minority | majority accuracy: 60.10 -> 59.84 | No clear effect |
| add FedReLa to FedYoYo | FedYoYo · ImageNet-LT alpha=0.1 | overall accuracy: 38.15 -> 38.78 | Improved |
| add FedReLa to FedLoGe | FedLoGe · ImageNet-LT alpha=0.1 | overall accuracy: 30.52 -> 31.70 | Improved |
| use z-score + tanh instead of min-max normalization for re-labeling probability (ablation) | FedETF + FedReLa · CIFAR-100-LT IF=100 | tail accuracy: 21.46 -> 24.52 | Improved |
| use z-score + tanh instead of min-max normalization (ablation) | FedETF + FedReLa · CIFAR-100-LT IF=100 | overall accuracy: 43.33 -> 44.54 | Improved |
| add FedReLa to FedETF as a baseline (ablation) | FedETF · CIFAR-100-LT IF=100 | overall accuracy: 42.81 -> 43.33 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FedLOGE — FedLoGe: Joint Local and Generic Federated Learning under Long-tailed Data (ICLR 2024) · better on 7 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Combating Data Imbalances in Federated Semi-supervised Learning with Dual Regulators (AAAI 2024 · similarity 0.90)
- Exploring Vacant Classes in Label-Skewed Federated Learning (AAAI 2025 · similarity 0.90)
- FedWon: Triumphing Multi-domain Federated Learning Without Normalization (ICLR 2024 · similarity 0.89)
- (FL)$^2$: Overcoming Few Labels in Federated Semi-Supervised Learning (NeurIPS 2024 · similarity 0.89)
- FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts (NeurIPS 2025 · similarity 0.88)
- FedLMT: Tackling System Heterogeneity of Federated Learning via Low-Rank Model Training with Theoretical Guarantees (ICML 2024 · 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.
