Keywords: Federated Learning, Malfunctioning Client Detection, Medical Image Classification, Outlier Detection
Abstract: Federated Learning (FL) allows the training of deep neural networks in a distributed and
privacy-preserving manner. However, this concept suffers from malfunctioning updates
sent by the attending clients that cause global model performance degradation. Reasons
for this malfunctioning might be technical issues, disadvantageous training data, or mali-
cious attacks. Most of the current defense mechanisms are meant to require impractical
prerequisites like knowledge about the number of malfunctioning updates, which makes
them unsuitable for real-world applications. To counteract these problems, we introduce
a novel method called ASMR, that dynamically excludes malfunctioning clients based on
their angular distance. Our novel method does not require any hyperparameters or knowl-
edge about the number of malfunctioning clients. Our experiments showcase the detection
capabilities of ASMR in an image classification task on a histopathological dataset, while
also presenting findings on the significance of dynamically adapting decision boundaries.
Latex Code: zip
Copyright Form: pdf
Submission Number: 113
Loading