Research map / Federated learning and privacy / Communication-efficient FL
DELTA: Diverse Client Sampling for Fasting Federated Learning
Lin Wang, Yongxin Guo, Tao Lin, Xiaoying Tang
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 gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · FashionMNIST | accuracy (%): 71.69 ±0.43 -> 72.10 ±0.49 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs Cluster-based IS · FashionMNIST | rounds to 70% accuracy: 362 -> 322 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs Cluster-based IS · FashionMNIST | time to 70% accuracy (s): 1547.41 -> 1372.33 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CIFAR-10 | accuracy (%): 55.05 ±0.27 -> 55.20 ±0.26 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CIFAR-10 | rounds to 54% accuracy: 313 -> 303 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CIFAR-10 | time to 54% accuracy (s): 3085.05 -> 2989.98 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · FEMNIST | accuracy (%): 80.11 ±0.29 -> 81.44 ±0.28 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · FEMNIST | rounds to 70% accuracy: 110 -> 98 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · FEMNIST | time to 70% accuracy (s): 223.27 -> 198.95 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CelebA | accuracy (%): 88.12 ±0.71 -> 89.67 ±0.56 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CelebA | rounds to 85% accuracy: 327 -> 306 | Improved |
| Replace gradient-norm importance sampling with diversity-and-local-variance sampling | FedPracDELTA vs FedPracIS · CelebA | time to 85% accuracy (s): 2746.82 -> 2607.12 | Improved |
Closest papers by meaning
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 2025 · similarity 0.91)
- Heterogeneity-Guided Client Sampling: Towards Fast and Efficient Non-IID Federated Learning (NeurIPS 2024 · similarity 0.90)
- FedAdaVR: Adaptive Variance Reduction for Robust Federated Learning under Limited Client Participation (NeurIPS 2026 · similarity 0.90)
- 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)
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable (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.
