Isolation Distributional Kernel: A New Tool for Point and Group Anomaly Detections
Lists
Only you see this
Abstract
We introduce Isolation Distributional Kernel as a new way to measure the similarity between two distributions. Existing approaches based on kernel mean embedding, which convert a point kernel to a distributional kernel, have two key issues: the point kernel employed has a feature map with intractable dimensionality; and it is <i>data independent</i> . This paper shows that Isolation Distributional Kernel (IDK), which is based on a <i>data dependent</i> point kernel, addresses both key issues. We demonstrate IDK’s efficacy and efficiency as a new tool for kernel-based anomaly detection for both point and group anomalies. Without explicit learning, using IDK alone outperforms existing kernel-based point anomaly detector OCSVM and other kernel mean embedding methods that rely on Gaussian kernel. For group anomaly detection, we introduce an IDK based detector called IDK <inline-formula><tex-math notation="LaTeX">$^2$</tex-math></inline-formula> . It reformulates the problem of group anomaly detection in input space into the problem of point anomaly detection in Hilbert Space, without the need for learning. IDK <inline-formula><tex-math notation="LaTeX">$^2$</tex-math></inline-formula> runs orders of magnitude faster than group anomaly detector OCSMM. We reveal for the first time that an effective kernel-based anomaly detector based on kernel mean embedding must employ a characteristic kernel which is data dependent.
BibTeX
@article{ting2023isolation,
author = {Kai Ming Ting and Bi-Cun Xu and Takashi Washio and Zhi-Hua Zhou},
title = {{Isolation Distributional Kernel: A New Tool for Point and Group Anomaly Detections}},
journal = {IEEE Trans. Knowl. Data Eng.},
year = {2023},
url = {https://openreview.net/forum?id=xK7AK6w8pMd},
}Loading related content...