Research map / Federated learning and privacy / Communication-efficient FL
Resilience Beyond Stationary Client Unavailability: Unlocking Efficient and Unbiased Federated Learning
Ming Xiang, Stratis Ioannidis, Edmund Yeh, Carlee Joe-Wong, Lili Su
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 interpolation (k=100) instead of k=0 (ablation) | FedSWE · CIFAR-10, stationary, P_delta=1 | test accuracy (%): 66.3 -> 66.9 | Improved |
| add interpolation (k=100) instead of k=0 (ablation) | FedSWE · CINIC-10, stationary, P_delta=1 | test accuracy (%): 47.3 -> 47.9 | Improved |
| add interpolation (k=100) instead of k=0 (ablation) | FedSWE · SVHN, stationary, P_delta=1 | test accuracy (%): 86.1 -> 85.4 | No clear effect |
| use FedSWE instead of FedAvg over active clients | FedAvg · SVHN, stationary, P_delta=1 | test accuracy (%): 82.4 -> 85.4 | Improved |
| use FedSWE instead of FedAvg over active clients | FedAvg · CIFAR-10, stationary, P_delta=1 | test accuracy (%): 62.9 -> 66.9 | Improved |
| add interpolation (k>0) under periodic unavailability (ablation) | FedSWE · SVHN and CIFAR-10, P_delta=100 | consensus error | Improved |
| increase interpolation k toward m=100 (ablation) | FedSWE · SVHN and CIFAR-10, P_delta=100 | rounds to 40% test accuracy | Improved |
| increase interpolation k beyond m=100 (ablation) | FedSWE · SVHN and CIFAR-10, P_delta=100 | rounds to 40% test accuracy | Hurt |
| increase interpolation k up to m then beyond (ablation) | FedSWE · SVHN and CIFAR-10, P_delta=100 | final test accuracy | Mixed |
| use FedSWE with k=100 instead of FedKnown | FedKnown · CIFAR-10, P_delta=100 | test accuracy | Improved |
| use FedSWE with k=100 instead of FedKnown | FedKnown · SVHN, P_delta=100 | test accuracy | No clear effect |
Closest papers by meaning
- Efficient Federated Learning against Heterogeneous and Non-stationary Client Unavailability (NeurIPS 2024 · similarity 0.96)
- FedAdaVR: Adaptive Variance Reduction for Robust Federated Learning under Limited Client Participation (NeurIPS 2026 · similarity 0.88)
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable (NeurIPS 2025 · similarity 0.88)
- Communication-Efficient Federated Learning with Accelerated Client Gradient (CVPR 2024 · similarity 0.88)
- Federated Learning under Periodic Client Participation and Heterogeneous Data: A New Communication-Efficient Algorithm and Analysis (NeurIPS 2024 · similarity 0.88)
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 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.
