Research map / Federated learning and privacy / Communication-efficient FL
Adaptive Client Sampling in Federated Learning via Online Learning with Bandit Feedback
Boxin Zhao, Lingxiao Wang, Ziqi Liu, Zhiqiang Zhang, Jun Zhou, Chaochao Chen, Mladen Kolar
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 |
|---|---|---|---|
| use Adaptive-OSMD Sampler instead of uniform sampling under high client heterogeneity (sigma=10.0) | mini-batch SGD · Simulated linear regression, M=100 clients | training loss and cumulative regret | Improved |
| use Adaptive-OSMD Sampler instead of uniform sampling under low heterogeneity (sigma=1.0) | mini-batch SGD · Simulated linear regression, M=100 clients | training loss | No clear effect |
| use Adaptive-OSMD Sampler instead of MABS, VRB and Avare bandit samplers | mini-batch SGD · Simulated linear regression, M=100 clients | training loss and cumulative regret | Improved |
| vary the mixing parameter alpha in {0.01,0.1,0.4,0.7,0.9,1.0} (ablation) | Adaptive-OSMD Sampler · Simulated linear regression, M=100 clients | training loss and cumulative regret | No clear effect |
| set alpha=1.0, which makes the sampler output a uniform distribution (ablation) | Adaptive-OSMD Sampler · Simulated linear regression, M=100 clients | training loss | Hurt |
| use dynamic Adaptive-OSMD distribution instead of fixed Lipschitz-based importance sampling | mini-batch SGD · Simulated sparse-feature regression, varying nu | training loss | Improved |
| use Adaptive-OSMD Sampler instead of uniform sampling and online samplers with skewed client sizes | FedAvg-style SGD with logistic regression · MNIST, KMNIST, FMNIST (500 clients, K=10) | training loss and validation accuracy | Improved |
Closest papers by meaning
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable (NeurIPS 2025 · similarity 0.89)
- Heterogeneity-Guided Client Sampling: Towards Fast and Efficient Non-IID Federated Learning (NeurIPS 2024 · similarity 0.88)
- Adaptive Federated Learning with Auto-Tuned Clients (ICLR 2024 · similarity 0.88)
- Pursuing Overall Welfare in Federated Learning through Sequential Decision Making (ICML 2024 · similarity 0.88)
- Communication-Efficient Federated Non-Linear Bandit Optimization (ICLR 2024 · similarity 0.87)
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 2025 · similarity 0.87)
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.
