Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networks
Preprint
Abstract
One major goal of the AI security community is to securely and reliably produce and deploy deep learning models for real-world applications. To this end, data poisoning based backdoor attacks on deep neural networks (DNNs) in the production stage (or training stage) and corresponding defenses are extensively explored in recent years. Ironically, backdoor attacks in the deployment stage, which can often happen in unprofessional users' devices and are thus arguably far more threatening in real-world scenarios, draw much less attention of the community. We attribute this imbalance of vigilance to the weak practicality of existing deployment-stage backdoor attack algorithms and the insufficiency of real-world attack demonstrations. To fill the blank, in this work, we study the realistic threat of deployment-stage backdoor attacks on DNNs. We base our study on a commonly used deployment-stage attack paradigm -- adversarial weight attack, where adversaries selectively modify model weights to embed backdoor into deployed DNNs. To approach realistic practicality, we propose the first gray-box and physically realizable weights attack algorithm for backdoor injection, namely subnet replacement attack (SRA), which only requires architecture information of the victim model and can support physical triggers in the real world. Extensive experimental simulations and system-level real-world attack demonstrations are conducted. Our results not only suggest the effectiveness and practicality of the proposed attack algorithm, but also reveal the practical risk of a novel type of computer virus that may widely spread and stealthily inject backdoor into DNN models in user devices. By our study, we call for more attention to the vulnerability of DNNs in the deployment stage.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Model Agnostic Defence Against Backdoor Attacks in Machine LearningIEEE Transactions on Reliability, 2022
- Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and DefensesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Backdoor Pre-trained Models Can Transfer to All2021
- T-BFA: Targeted Bit-Flip Adversarial Weight AttackIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- Detecting AI Trojans Using Meta Neural Analysis2021
- Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbation2020
- A Backdoor Attack Against LSTM-Based Text Classification SystemsIEEE Access, 2019
- Adversarial Examples in the Physical World2018
- Flipping bits in memory without accessing them: An experimental study of DRAM disturbance errors2014
- Possible macro virus attacks and how to prevent themComputers & Security, 1996