Unveiling Neural Network Data Free Backdoor Threats in Industrial Control Systems
Zijian Zhang, Isra Elsharef, Zhen Zeng · 2023
The neural network data-free backdoor attack is an emerging and potent threat, which requires minimal resources and does not rely on original training data to implant backdoors. This threat poses a significant risk to industrial control systems, where increasing integration of neural network applications within systems heightens their vulnerabilities. This study represents preliminary work in understanding the data-free backdoor attack and assessing its potential impact on industrial control systems (ICSs). The key factors influencing attack performance are examined through experimental research. Potential risks specific to ICS are identified through threat modeling. The insights gained could inform the development of more robust defense strategies, enhancing the protection of critical ICS infrastructure against these neural network backdoor threats.