Published in , 2025
Z. Jiang, X. Lyu, S. Shi, Y. Xiao, Y. Chen, Y. Hou, W. Lou, N. Wang ‘28th European Conference on Artificial Intelligence’ pdf
Federated learning is vulnerable to both data poisoning and model poisoning attacks. My research develops robust defenses to identify malicious clients while preserving collaborative learning.
Adversarial attacks found in computer vision apply to network intrusion detection systems (IDS). We analyze multiple AE methods and discover they share one common characteristic: AE is close to the manifold of the true class rather than the target class because of the small perturbation size. Most AEs maintain a small perturbation size to maintain the original property and to hide the maliciousness. We design a detection rule as any input is an AE if there is an INCONSISTENCY between manifold evaluation and model classification.
Federated learning is a promising framework for healthcare, etc. FL provides privacy protection by its design as data remains on a local device. However, the model parameter still leaks privacy. In order to protect local data privacy, we applied differential privacy by carefully adding random noise to model parameters. To minimize model accuracy degradation, we propose an adaptive clipping method to adaptively add noise to gradients according to the change of gradient scale. Our methods have been proven to improve model accuracy without degrading privacy protection.
Machine learning (ML) has shown advances in network intrusion detection systems because of its capability to detect zero-day attacks. However, ML-based intrusion detection usually suffers from high false positives compared with the traditional signature-based network intrusion detection system. To reduce both false positives and false negatives, we developed a contrastive learning-based intrusion detection system. The proposed detection mechanism extracted the key common properties of benign variations to learn a more accurate model for the benign class. To apply our detection model to IoT systems and alleviate privacy concerns, we incorporated FEderated learning framework into the Contrastive-learning-based detection method and proposed a system, FeCo. FeCo significantly reduced false positives and improved intrusion detection accuracy compared to previous works. Through extensive experiments on the NSL-KDD and BaIoTdatasets, we demonstrated that FeCoachieves a large accuracy improvement (as high as 8%) compared to the state-of-the-art detection methods.
Published:
This talk presents FeCo, a machine-learning-based IDS for IoT networks. FeCo incorporates contrastive learning into FL framework to support distributed intrusion detection. FeCo obtains more representative feature vectors by contrastive learning. These feature vectors have higher discriminative power between normal and malicious traffic. This effectively enables FeCo to achieve better detection accuracy than other baselines. Through extensive evaluations on the NSL-KDD dataset, we demonstrate the high effectiveness of FeCo in both centralized and federated learning setting.
Published:
This talk presents a robust aggregation algorithm FLARE to protect FL against MPAs. It demonstrates that PLR vector has high potentials in differentiating malicious/poisonous models from the benign ones. FLARE effectively minimizes the impact of malicious/poisonous models on the final aggregation by assigning low trust scores to those with diverging PLRs.
Undergraduate course, University of South Florida, Department of Computer Science and Engineering, 2024
2024 Spring, 2025 Spring.
Graduate course, University of South Florida, Department of Computer Science and Engineering, 2024
This course is created by Dr. Wang. 2023 Fall, 2024 Fall
Undergraduate/Graduate course, University of South Florida, Bellini College of AI, Cybersecurity and Computing, 2026
2026 Spring.