FedPara: Low-Rank Hadamard Product for Communication-Efficient Federated Learning

Abstract

In this work, we propose a communication-efficient parameterization, FedPara, for federated learning (FL) to overcome the burdens on frequent model uploads and downloads. Our method re-parameterizes weight parameters of layers using low-rank weights followed by the Hadamard product. Compared to the conventional low-rank parameterization, our FedPara method is not restricted to low-rank constraints, and thereby it has a far larger capacity. This property enables to achieve comparable performance while requiring 3 to 10 times lower communication costs than the model with the original layers, which is not achievable by the traditional low-rank methods. The efficiency of our method can be further improved by combining with other efficient FL optimizers. In addition, we extend our method to a personalized FL application, pFedPara, which separates parameters into global and local ones. We show that pFedPara outperforms competing personalized FL methods with more than three times fewer parameters.

arXiv commentAccepted at ICLR 2022

References in corpus (15)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Communication-Efficient Learning of Deep Networks from Decentralized Data
• Federated Learning: Strategies for Improving Communication Efficiency
• Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
• QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding
• TernGrad: Ternary Gradients to Reduce Communication in Distributed Deep Learning
• Federated Learning with Personalization Layers
• FedML: A Research Library and Benchmark for Federated Machine Learning
• FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization
• LEAF: A Benchmark for Federated Settings
• A Survey of Privacy Attacks in Machine Learning
• HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clients
• PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization
• Federated Learning Based on Dynamic Regularization
• Federated Learning with Only Positive Labels