Federated Learning with Matched Averaging
arXiv:2002.06440
Abstract
Federated learning allows edge devices to collaboratively learn a shared model while keeping the training data on device, decoupling the ability to do model training from the need to store the data in the cloud. We propose Federated matched averaging (FedMA) algorithm designed for federated learning of modern neural network architectures e.g. convolutional neural networks (CNNs) and LSTMs. FedMA constructs the shared global model in a layer-wise manner by matching and averaging hidden elements (i.e. channels for convolution layers; hidden states for LSTM; neurons for fully connected layers) with similar feature extraction signatures. Our experiments indicate that FedMA not only outperforms popular state-of-the-art federated learning algorithms on deep CNN and LSTM architectures trained on real world datasets, but also reduces the overall communication burden.
Accepted by ICLR 2020
References in corpus (5)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- On the Convergence of Adam and Beyond
- Towards Federated Learning at Scale: System Design
- No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing World
- Bayesian Nonparametric Federated Learning of Neural Networks