Corpus ID: 211678094

Adaptive Federated Optimization

@article{Reddi2020AdaptiveFO,
  title={Adaptive Federated Optimization},
  author={Sashank J. Reddi and Zachary B. Charles and M. Zaheer and Zachary A. Garrett and Keith Rush and Jakub Konecn{\'y} and S. Kumar and H. B. McMahan},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.00295}
}
Federated learning is a distributed machine learning paradigm in which a large number of clients coordinate with a central server to learn a model without sharing their own training data. Due to the heterogeneity of the client datasets, standard federated optimization methods such as Federated Averaging (FedAvg) are often difficult to tune and exhibit unfavorable convergence behavior. In non-federated settings, adaptive optimization methods have had notable success in combating such issues. In… Expand
63 Citations
Mime: Mimicking Centralized Stochastic Algorithms in Federated Learning.
  • 12
  • PDF
FedCluster: Boosting the Convergence of Federated Learning via Cluster-Cycling
  • 3
  • PDF
Double Momentum SGD for Federated Learning
  • PDF
FedADC: Accelerated Federated Learning with Drift Control
  • 1
  • PDF
Fast-Convergent Federated Learning
  • 10
  • PDF
Semi-Synchronous Federated Learning
  • 1
  • PDF
CosSGD: Nonlinear Quantization for Communication-efficient Federated Learning
  • PDF
FLaPS: Federated Learning and Privately Scaling
  • PDF
Communication-Efficient Federated Learning via Optimal Client Sampling
  • 4
  • PDF
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 50 REFERENCES
Federated Optimization in Heterogeneous Networks
  • 301
  • PDF
SCAFFOLD: Stochastic Controlled Averaging for Federated Learning
  • 85
  • PDF
LEAF: A Benchmark for Federated Settings
  • 186
  • PDF
Adaptive Federated Learning in Resource Constrained Edge Computing Systems
  • 371
  • PDF
Federated Learning: Challenges, Methods, and Future Directions
  • 454
  • PDF
Towards Federated Learning at Scale: System Design
  • 584
  • PDF
On the Convergence of FedAvg on Non-IID Data
  • 242
  • PDF
Communication-Efficient Learning of Deep Networks from Decentralized Data
  • 2,061
  • PDF
...
1
2
3
4
5
...