Research Area:  Machine Learning
Since deep neural networks can learn data representation from training data automatically, deep learning methods are widely used in the network anomaly detection. However, challenges of deep learning-based anomaly detection methods still exist, the major of which is the training data scarcity problem. In this paper, we propose a novel network anomaly detection method (NAFT) using federated learning and transfer learning to overcome the data scarcity problem. In the first learning stage, a people or organization 𝑂𝑡, who intends to conduct a detection model for a specific attack, can join in the federated learning with a similar training task to learn basic knowledge from other participants training data. In the second learning stage, 𝑂𝑡 uses the transfer learning method to reconstruct and re-train the model to further improve the detection performance on the specific task. Experiments conducted on the UNSW-NB15 dataset show that the proposed method can achieve a better anomaly detection performance than other baseline methods when training data is scarce.
Keywords:  
Author(s) Name:  Ying Zhao, Junjun Chen, Qianling Guo, Jian Teng & Di Wu
Journal name:  
Conferrence name:  Security and Privacy in Digital Economy
Publisher name:  Springer
DOI:  10.1007/978-981-15-9129-7_16
Volume Information:  
Paper Link:   https://link.springer.com/chapter/10.1007/978-981-15-9129-7_16