Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedSR: A Simple and Effective Domain Generalization Method for Federated Learning
A. Tuan Nguyen, Philip Torr, Ser-Nam Lim
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 L2R and CMI regularizers on representation to FedAVG (FedSR) | FedAVG · PACS | average accuracy: 80.3 -> 83.7 | Improved |
| add L2R and CMI regularizers to FedAVG | FedAVG · DomainNet | average accuracy: 39.6 -> 41.3 | Improved |
| add L2R and CMI regularizers to FedAVG | FedAVG · OfficeHome | average accuracy: 67.9 -> 69.3 | Improved |
| add L2R and CMI regularizers to FedAVG | FedAVG · RotatedMNIST | average accuracy: 94.5 -> 96.7 | Improved |
| add CMI regularizer on top of L2R (FedSR vs FedL2R) (ablation) | FedSR · PACS | average accuracy: 83.1 -> 83.7 | Improved |
| add CMI regularizer on top of L2R (FedSR vs FedL2R) (ablation) | FedSR · OfficeHome | average accuracy: 68.8 -> 69.3 | Improved |
| add CMI regularizer on top of L2R (FedSR vs FedL2R) (ablation) | FedSR · DomainNet | average accuracy: 40.6 -> 41.3 | Improved |
| add L2R on top of CMI (FedSR vs FedCMI) (ablation) | FedSR · PACS | average accuracy: 81.7 -> 83.7 | Improved |
| FedCMI alone vs FedAVG (ablation) | FedAVG · OfficeHome | average accuracy: 67.9 -> 67.8 | No clear effect |
| FedL2R alone vs FedAVG (ablation) | FedAVG · PACS | average accuracy: 80.3 -> 83.1 | Improved |
Later papers that compared with this method
7 later accepted papers report FedSR as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.
- Deploying Models to Non-participating Clients in Federated Learning without Fine-tuning: A Hypernetwork-based Approach (ICLR 2026 · better on 1 of 1 shared result)
- Causality Inspired Federated Learning for OOD Generalization (ICML 2025 · better on 5 of 5 shared results)
- Beyond the Federation: Topology-aware Federated Learning for Generalization to Unseen Clients (ICML 2024 · better on 2 of 2 shared results)
- DapperFL: Domain Adaptive Federated Learning with Model Fusion Pruning for Edge Devices (NeurIPS 2024 · better on 2 of 2 shared results)
- Local and Global Flatness for Federated Domain Generalization (ECCV 2024 · better on 1 of 1 shared result)
- Out-of-Distribution Generalization of Federated Learning via Implicit Invariant Relationships (ICML 2023 · better on 10 of 10 shared results)
- StableFDG: Style and Attention Based Learning for Federated Domain Generalization (NeurIPS 2023 · better on 4 of 4 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
- Federated Domain Generalization with Data-free On-server Matching Gradient (ICLR 2025 · similarity 0.91)
- FedWon: Triumphing Multi-domain Federated Learning Without Normalization (ICLR 2024 · similarity 0.91)
- Generalization in Federated Learning: A Conditional Mutual Information Framework (ICML 2025 · similarity 0.90)
- Efficiently Assemble Normalization Layers and Regularization for Federated Domain Generalization (CVPR 2024 · similarity 0.90)
- Benchmarking Algorithms for Federated Domain Generalization (ICLR 2024 · similarity 0.89)
- FedLMT: Tackling System Heterogeneity of Federated Learning via Low-Rank Model Training with Theoretical Guarantees (ICML 2024 · similarity 0.89)
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.
