Research map / Federated learning and privacy / Personalized and heterogeneous FL
Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions
Hossein Zakerinia, Jonathan Scott, Christoph H. Lampert
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 unlabeled clients to training (ablation) | FLowDUP (CNN) · CIFAR10 p=0.1 | accuracy: 65.3 -> 67.5 | Improved |
| add unlabeled clients to training (ablation) | FLowDUP (CNN) · CIFAR10 p=0.5 | accuracy: 82.5 -> 83.8 | No clear effect |
| add unlabeled clients to training (ablation) | FLowDUP (ResNet18) · CIFAR10 p=0.2 | accuracy: 83.3 -> 83.0 | No clear effect |
| add unlabeled clients to training (ablation) | FLowDUP (ResNet18) · CIFAR10 p=0.05 | accuracy: 54.8 -> 57.5 | Improved |
| learn the regularizer psi_r instead of fixing it to 0 (ablation) | FLowDUP (CNN) · CIFAR10 p=1.0 | accuracy: 85.6 -> 86.6 | Improved |
| learn the regularizer psi_r instead of fixing it to 0 (ablation) | FLowDUP (CNN) · CIFAR10 p=0.1 | accuracy: 65.5 -> 67.5 | Improved |
| increase subspace dimension k from 200 to 10000 (ablation) | FLowDUP (CNN) · CIFAR10 p=1.0 | accuracy: 71.2 -> 86.6 | Improved |
| increase subspace dimension k from 200 to 10000 (ablation) | FLowDUP (CNN) · CIFAR10 p=0.1 | accuracy: 56.6 -> 66.1 | Improved |
| use ResNet18 instead of CNN as client model f (h1 = CNN) (ablation) | FLowDUP · CIFAR10 p=0.1 | accuracy: 66.1 -> 71.4 | Improved |
| use ResNet18 instead of CNN as hypernetwork encoder h1 (f = CNN) (ablation) | FLowDUP · CIFAR10 p=0.5 | accuracy: 83.3 -> 87.3 | Improved |
| personalize with FLowDUP instead of FedTTA | FLowDUP vs FedTTA · FEMNIST p=1.0 | accuracy: 74.3 -> 89.3 | Improved |
| use FLowDUP instead of FedAvg | FLowDUP vs FedAvg · CIFAR10 CNN p=0.1 | accuracy: 47.9 -> 66.1 | Improved |
Closest papers by meaning
- Personalized Federated Learning With Gaussian Processes (NeurIPS 2021 · similarity 0.88)
- pFedClub: Controllable Heterogeneous Model Aggregation for Personalized Federated Learning (NeurIPS 2024 · similarity 0.88)
- PeFLL: Personalized Federated Learning by Learning to Learn (ICLR 2024 · similarity 0.86)
- FedImpro: Measuring and Improving Client Update in Federated Learning (ICLR 2024 · similarity 0.86)
- A Bayesian Approach for Personalized Federated Learning in Heterogeneous Settings (NeurIPS 2024 · similarity 0.86)
- FedAS: Bridging Inconsistency in Personalized Federated Learning (CVPR 2024 · similarity 0.85)
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.
