Detecting Poisoning Attacks on Federated Learning Using Gradient-Weighted Class Activation Mapping
Jingjing Zheng, Kai Li, Xin Yuan, Wei Ni, Eduardo Tovar
Abstract
This paper proposes a new defense mechanism, namely, GCAMA, against model poisoning attacks on Federated learning (FL), which integrates <u>G</u>radient-weighted <u>C</u>lass <u>A</u>ctivation <u>M</u>apping (GradCAM) and <u>A</u>utoencoder to offer a scientifically more powerful detection capability compared to existing Euclidean distance-based approaches. Particularly, GCAMA generates a heat map for each uploaded local model update, transforming each local model update into a lower-dimensional, visual representation, thereby accentuating the hidden features of the heat maps and increasing the success rate of identifying anomalous heat maps and malicious local models. We test ResNet-18 and MobileNetV3-Large deep learning models with CIFAR-10 and GTSRB datasets under Non-Independent and Identically Distributed (Non-IID) setting, respectively. The results demonstrate that GCAMA offers superior test accuracy of FL global model compared to the state-of-the-art methods. Our code is available at: https://github.com/jjzgeeks/GradCAM-AE