Research map / Federated learning and privacy / Communication-efficient FL
FedDO: Dynamic Client Optimization for Adaptive Federated Learning
Peixuan Tang, Xuehe Wang
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 |
|---|---|---|---|
| replace FedAvg with FedDO RL data-usage control | FedAvg · CIFAR-10 CNN α=0.5 | accuracy: 56.88 -> 89.70 | Improved |
| replace FedAvg with FedDO | FedAvg · TinyImageNet CNN α=0.1 | accuracy: 12.01 -> 32.45 | Improved |
| replace FedAvg with FedDO | FedAvg · CIFAR-100 CNN α=0.1 | accuracy: 28.92 -> 50.14 | Improved |
| replace FedAvg with FedDO | FedAvg · CIFAR-10 ResNet-18 IID | accuracy: 66.07 -> 77.03 | Improved |
| replace FedLC with FedDO | FedLC · CIFAR-10 VGG-11 α=0.1 | accuracy: 88.55 -> 81.76 | Hurt |
| replace FedAvg with FedDO | FedAvg · CIFAR-10 ResNet-18 α=0.5 | accuracy: 87.08 -> 88.26 | Improved |
| replace FedAvg with FedDO | FedAvg · FashionMNIST CNN IID | accuracy: 90.80 -> 90.99 | No clear effect |
| replace FedAvg with FedDO | FedAvg · CIFAR-10 ResNet-18 Dir(0.5) | communication rounds to target: 784 -> 429 | Improved |
| use FedDO instead of other baselines | FedCross · CIFAR-10 ResNet-18 Dir(0.1) | communication rounds to target: 729 -> 344 | Improved |
| add low-rank parameterization to agent, K=100 (ablation) | DSAC-T agent · FashionMNIST VGG-11 IID | accuracy: 90.62 -> 90.29 | Hurt |
| add low-rank parameterization to agent, K=500 (ablation) | DSAC-T agent · FashionMNIST VGG-11 IID | accuracy: 91.03 -> 89.27 | Hurt |
| add low-rank parameterization to agent, K=1000 (ablation) | DSAC-T agent · FashionMNIST VGG-11 IID | accuracy: 89.78 -> 89.55 | Hurt |
Closest papers by meaning
- FedECADO: A Dynamical System Model of Federated Learning (ICML 2025 · similarity 0.92)
- Adaptive Federated Learning with Auto-Tuned Clients (ICLR 2024 · similarity 0.90)
- Efficient Adaptive Federated Optimization (NeurIPS 2025 · similarity 0.90)
- FedAdaVR: Adaptive Variance Reduction for Robust Federated Learning under Limited Client Participation (NeurIPS 2026 · similarity 0.89)
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable (NeurIPS 2025 · similarity 0.89)
- FedSMU: Communication-Efficient and Generalization-Enhanced Federated Learning through Symbolic Model Updates (ICML 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.
