FL-MGVN: Federated learning for anomaly detection using mixed gaussian variational self-encoding network

作者：

Highlights：

• Propose an anomaly detection classification model that incorporates federated learning and mixed Gaussian variational self-coding networks.

• Realize anomaly detection based on federated learning, including network attack and sample dissimilarity.

• The proposed MGVN network model first constructs a variational self-coder using a mixed gaussian prior to extract features from the input data, and then constructs a deep support vector network with a mixed gaussian variational self-coder.

• Verify the multi-classification anomaly detection performance on benchmark datasets such as NSL-KDD, MNIST and Fashion-MNIST.

摘要

•Propose an anomaly detection classification model that incorporates federated learning and mixed Gaussian variational self-coding networks.•Realize anomaly detection based on federated learning, including network attack and sample dissimilarity.•The proposed MGVN network model first constructs a variational self-coder using a mixed gaussian prior to extract features from the input data, and then constructs a deep support vector network with a mixed gaussian variational self-coder.•Verify the multi-classification anomaly detection performance on benchmark datasets such as NSL-KDD, MNIST and Fashion-MNIST.

论文关键词：Federated learning,Anomaly detection,Deep learning,Variational self-encoder,Mixed gaussian distribution

论文评审过程：Received 3 September 2021, Revised 22 November 2021, Accepted 23 November 2021, Available online 20 December 2021, Version of Record 20 December 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102839