Research map / Federated learning and privacy / Communication-efficient FL
FedBR: Improving Federated Learning on Heterogeneous Data via Local Learning Bias Reduction
Yongxin Guo, Xiaoying Tang, Tao Lin
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 classifier-output balancing to FedAvg (ablation) | FedAvg · RotatedMNIST | accuracy (%): 82.47 -> 84.40 | Improved |
| Add classifier-output balancing to FedAvg (ablation) | FedAvg · CIFAR10 | accuracy (%): 58.99 -> 64.32 | Improved |
| Add classifier-output balancing to FedAvg (ablation) | FedAvg · CIFAR100 | accuracy (%): 46.37 -> 47.22 | Improved |
| Add adversarial feature debiasing to FedAvg (ablation) | FedAvg · RotatedMNIST | accuracy (%): 82.47 -> 86.25 | Improved |
| Add adversarial feature debiasing to FedAvg (ablation) | FedAvg · CIFAR10 | accuracy (%): 58.99 -> 63.44 | Improved |
| Add adversarial feature debiasing to FedAvg (ablation) | FedAvg · CIFAR100 | accuracy (%): 46.37 -> 47.78 | Improved |
| Add the max step to adversarial feature debiasing (ablation) | Component 2 · RotatedMNIST | accuracy (%): 81.24 -> 86.25 | Improved |
| Add the max step to adversarial feature debiasing (ablation) | Component 2 · CIFAR10 | accuracy (%): 58.84 -> 63.44 | Improved |
| Add the max step to adversarial feature debiasing (ablation) | Component 2 · CIFAR100 | accuracy (%): 43.50 -> 47.78 | Improved |
| Combine classifier-output balancing with adversarial feature debiasing (ablation) | FedBR · RotatedMNIST | accuracy (%): Component 1: 84.40; Component 2: 86.25; FedBR: 86.58 | Improved |
| Combine classifier-output balancing with adversarial feature debiasing (ablation) | FedBR · CIFAR10 | accuracy (%): Component 1: 64.32; Component 2: 63.44; FedBR: 64.65 | Improved |
| Combine classifier-output balancing with adversarial feature debiasing (ablation) | FedBR · CIFAR100 | accuracy (%): Component 1: 47.22; Component 2: 47.78; FedBR: 47.75 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports FedBR as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Incentives in Federated Learning: Equilibria, Dynamics, and Mechanisms for Welfare Maximization (NeurIPS 2023 · better on 10 of 10 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
- Low Precision Local Training is Enough for Federated Learning (NeurIPS 2024 · similarity 0.91)
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 2025 · similarity 0.89)
- FedLoGe: Joint Local and Generic Federated Learning under Long-tailed Data (ICLR 2024 · similarity 0.89)
- (FL)$^2$: Overcoming Few Labels in Federated Semi-Supervised Learning (NeurIPS 2024 · similarity 0.89)
- TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients (AAAI 2024 · similarity 0.88)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 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.
