Abstract

Federated learning enables resource-constrained edge compute devices, such as mobile phones and IoT devices, to learn a shared model for prediction, while keeping the training data local. This decentralized approach to train models provides privacy, security, regulatory and economic benefits. In this work, we focus on the statistical challenge of federated learning when local data is non-IID. We first show that the accuracy of federated learning reduces significantly, by up to 55% for neural networks trained for highly skewed non-IID data, where each client device trains only on a single class of data. We further show that this accuracy reduction can be explained by the weight divergence, which can be quantified by the earth mover's distance (EMD) between the distribution over classes on each device and the population distribution. As a solution, we propose a strategy to improve training on non-IID data by creating a small subset of data which is globally shared between all the edge devices. Experiments show that accuracy can be increased by 30% for the CIFAR-10 dataset with only 5% globally shared data.

Keywords

Computer scienceTrainFocus (optics)Enhanced Data Rates for GSM EvolutionFederated learningDivergence (linguistics)Class (philosophy)Machine learningArtificial intelligenceData miningMobile deviceArtificial neural networkEdge devicePopulationCloud computing

Related Publications

Publication Info

Year
2018
Type
preprint
Citations
1894
Access
Closed

External Links

Download PDF (Free) View on DOI.org arXiv Semantic Scholar

Social Impact

Altmetric
PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

1894
OpenAlex
264
Influential

Cite This

Yue Zhao, Meng Li, Liangzhen Lai et al. (2018). Federated Learning with Non-IID Data. arXiv (Cornell University) . https://doi.org/10.48550/arxiv.1806.00582

Identifiers

DOI
10.48550/arxiv.1806.00582
arXiv
1806.00582

Data Quality

Data completeness: 84%