Sequential Adversarial Anomaly Detection with Deep Fourier Kernel

Shixiang Zhu, Henry Shaowu Yuchi, Minghe Zhang, Yao Xie

Abstract

We present a novel adversarial detector for the anomalous sequence when there are only one-class training samples. The detector is developed by finding the best detector that can discriminate against the worst-case, which statistically mimics the training sequences. We explicitly capture the dependence in sequential events using the marked point process with a deep Fourier kernel. The detector evaluates a test sequence and compares it with an optimal time-varying threshold, which is also learned from data. Using numerical experiments on simulations and real-world datasets, we demonstrate the superior performance of our proposed method.
BibTeX
@inproceedings{zhu2021sequential,
  author = {Shixiang Zhu and Henry Shaowu Yuchi and Minghe Zhang and Yao Xie},
  title = {{Sequential Adversarial Anomaly Detection with Deep Fourier Kernel}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2021},
  doi = {10.1109/icassp39728.2021.9414197},
}

Loading related content...

Figure