Research Article
Backdoor Learning on Siamese Networks Using Physical Triggers: FaceNet as a Case Study
@INPROCEEDINGS{10.1007/978-3-031-56580-9_17, author={Zeshan Pang and Yuyuan Sun and Shasha Guo and Yuliang Lu}, title={Backdoor Learning on Siamese Networks Using Physical Triggers: FaceNet as a Case Study}, proceedings={Digital Forensics and Cyber Crime. 14th EAI International Conference, ICDF2C 2023, New York City, NY, USA, November 30, 2023, Proceedings, Part I}, proceedings_a={ICDF2C}, year={2024}, month={4}, keywords={Backdoor learning Physical trigger Multi-task learning Siamese networks FaceNet}, doi={10.1007/978-3-031-56580-9_17} }- Zeshan Pang
Yuyuan Sun
Shasha Guo
Yuliang Lu
Year: 2024
Backdoor Learning on Siamese Networks Using Physical Triggers: FaceNet as a Case Study
ICDF2C
Springer
DOI: 10.1007/978-3-031-56580-9_17
Abstract
Deep learning models play an important role in many real-world applications, for example, in face recognition systems, Siamese networks have been widely used. Their security issues have attracted increasing attention and backdoor learning is an emerging research area that studies the security of deep learning models. However, few backdoor learning focuses on Siamese models. To address the problem, this paper proposes a backdoor learning method on Siamese networks using physical triggers. Inspired by multi-task learning, after poisoning the dataset, the pre-trained Siamese network is fine-tuned at the last linear layer with the guidance of two tasks: outputting correct embeddings of benign samples and reacting to the poison samples. The outputs of the two tasks are then added and normalized as the output of the model. Experiments show that using the typical Siamese network FaceNet as the target network, the attack success rate of our method reaches 99%, while the model accuracy on the benign dataset decreases by only 0.001%, which reveals the model security issue.
