Using adversarial images to improve outcomes of federated learning for non-IID data

Anastasiya Danilenka; Maria Ganzha; Marcin Paprzycki; Jacek Mańdziuk

Using adversarial images to improve outcomes of federated learning for non-IID data

Machine Learning 2022-06-17 v1

Authors: Anastasiya Danilenka , Maria Ganzha , Marcin Paprzycki , Jacek Mańdziuk

Abstract

One of the important problems in federated learning is how to deal with unbalanced data. This contribution introduces a novel technique designed to deal with label skewed non-IID data, using adversarial inputs, created by the I-FGSM method. Adversarial inputs guide the training process and allow the Weighted Federated Averaging to give more importance to clients with 'selected' local label distributions. Experimental results, gathered from image classification tasks, for MNIST and CIFAR-10 datasets, are reported and analyzed.

Keywords

federated learning non-iid data adversarial examples label skew weighted federated averaging image classification

Cite

@article{arxiv.2206.08124,
  title  = {Using adversarial images to improve outcomes of federated learning for non-IID data},
  author = {Anastasiya Danilenka and Maria Ganzha and Marcin Paprzycki and Jacek Mańdziuk},
  journal= {arXiv preprint arXiv:2206.08124},
  year   = {2022}
}

Related papers

View all related →