IEEE Intelligent Systems · 2024 · 16 citations · 12 references
ComputingComputer ScienceCybersecurityComputer SecurityData MiningMachine LearningFederated LearningSecure Multi-party ComputationAttack SuccessBackdoor TechniquesClusteringModel AggregationPrivacy-preserving Model Updates
This article proposes a novel scheme, SupRTE, to suppress backdoor injection in federated learning via robust trust evaluation, which effectively prevents malicious updates from infiltrating the model aggregation process. The robust trust evaluation process in SupRTE consists of two components: 1) behavior representation extractor, creating individual profiles for each client through multidimensional information; 2) trust scorer, measuring the discrepancies between malicious and benign clients as trust scores by utilizing grading and clustering strategies. According to these trust scores, SupRTE can dynamically adjust the weight of each participating client to effectively suppress the malicious backdoor injection. Remarkably, SupRTE can be easily deployed on the server without requiring any auxiliary information and is highly adaptable to various Non-IID scenarios. Extensive experiments over 3 datasets against 2 kinds of backdoor variants are conducted. Experimental results demonstrate that SupRTE can significantly reduce the attack success rate to below 2% with a minimal impact on the main task accuracy and outperforms the state-of-the-art defense methods.
12
Federated Learning: Challenges, Methods, and Future Directions
IEEE Signal Processing Magazine · 2020 · 4.6K citations · Full text
hdbscan: Hierarchical density based clustering
Leland McInnes, John Healy, Steve Astels · The Journal of Open Source Software · 2017 · 2.7K citations · Full text
FLTrust: Byzantine-robust Federated Learning via Trust Bootstrapping
Xiaoyu Cao, Minghong Fang, Jia Liu et al. · 2021 · 790 citations · Full text
Ensemble Distillation for Robust Model Fusion in Federated Learning
Tao Lin, Lingjing Kong, Sebastian U. Stich et al. · arXiv (Cornell University) · 2020 · 486 citations · Full text