Alert button

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

Add code
Bookmark button
Alert button
Jun 16, 2022
Anastasiya Danilenka, Maria Ganzha, Marcin Paprzycki, Jacek Mańdziuk

Figure 1 for Using adversarial images to improve outcomes of federated learning for non-IID data
Figure 2 for Using adversarial images to improve outcomes of federated learning for non-IID data
Figure 3 for Using adversarial images to improve outcomes of federated learning for non-IID data
Figure 4 for Using adversarial images to improve outcomes of federated learning for non-IID data

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

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.

View paper onarxiv icon

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon