Idioma:

Reverse engineering convolutional neural networks through side-channel information leaks

Hua, Weizhe ; Zhang, Zhiru ; Suh, G. Edward

Proceedings of the 55th Annual Design Automation Conference, 2018, p.1-6

Texto completo disponível

Citações Citado por

Título:
Reverse engineering convolutional neural networks through side-channel information leaks
Autor: Hua, Weizhe ; Zhang, Zhiru ; Suh, G. Edward
Assuntos: Computing methodologies ; Computing methodologies -- Machine learning ; Computing methodologies -- Machine learning -- Machine learning approaches ; Computing methodologies -- Machine learning -- Machine learning approaches -- Neural networks
É parte de: Proceedings of the 55th Annual Design Automation Conference, 2018, p.1-6
Descrição: A convolutional neural network (CNN) model represents a crucial piece of intellectual property in many applications. Revealing its structure or weights would leak confidential information. In this paper we present novel reverse-engineering attacks on CNNs running on a hardware accelerator, where an adversary can feed inputs to the accelerator and observe the resulting off-chip memory accesses. Our study shows that even with data encryption, the adversary can infer the underlying network structure by exploiting the memory and timing side-channels. We further identify the information leakage on the values of weights when a CNN accelerator performs dynamic zero pruning for off-chip memory accesses. Overall, this work reveals the importance of hiding off-chip memory access pattern to truly protect confidential CNN models.
Editor: New York, NY, USA: ACM
Idioma: Inglês

Voltar para lista de resultados