Backdoor Attack Against One-Class Sequential Anomaly Detection Models
Abstract
We propose BA-OCAD, a clean-label backdoor attack framework targeting one-class sequential anomaly detection models. By injecting stealthy triggers into training data, BA-OCAD enables adversaries to mislead anomaly detectors at inference while maintaining benign performance.
Type
Publication
In Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2024)
Citation
@inproceedings{cheng2024backdoor,
title={Backdoor attack against one-class sequential anomaly detection models},
author={Cheng, He and Yuan, Shuhan},
booktitle={Pacific-Asia Conference on Knowledge Discovery and Data Mining},
pages={262--274},
year={2024},
organization={Springer}
}
