Detection and Recovery Against Deep Neural Network Fault Injection Attacks Based on Contrastive Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916110211743744 |
|---|---|
| author | Wang, Chenan Zhao, Pu Wang, Siyue Lin, Xue |
| author_facet | Wang, Chenan Zhao, Pu Wang, Siyue Lin, Xue |
| contents | Deep Neural Network (DNN) models when implemented on executing devices as the inference engines are susceptible to Fault Injection Attacks (FIAs) that manipulate model parameters to disrupt inference execution with disastrous performance. This work introduces Contrastive Learning (CL) of visual representations i.e., a self-supervised learning approach into the deep learning training and inference pipeline to implement DNN inference engines with self-resilience under FIAs. Our proposed CL based FIA Detection and Recovery (CFDR) framework features (i) real-time detection with only a single batch of testing data and (ii) fast recovery effective even with only a small amount of unlabeled testing data. Evaluated with the CIFAR-10 dataset on multiple types of FIAs, our CFDR shows promising detection and recovery effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16766 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Detection and Recovery Against Deep Neural Network Fault Injection Attacks Based on Contrastive Learning Wang, Chenan Zhao, Pu Wang, Siyue Lin, Xue Machine Learning Artificial Intelligence Cryptography and Security Computer Vision and Pattern Recognition Deep Neural Network (DNN) models when implemented on executing devices as the inference engines are susceptible to Fault Injection Attacks (FIAs) that manipulate model parameters to disrupt inference execution with disastrous performance. This work introduces Contrastive Learning (CL) of visual representations i.e., a self-supervised learning approach into the deep learning training and inference pipeline to implement DNN inference engines with self-resilience under FIAs. Our proposed CL based FIA Detection and Recovery (CFDR) framework features (i) real-time detection with only a single batch of testing data and (ii) fast recovery effective even with only a small amount of unlabeled testing data. Evaluated with the CIFAR-10 dataset on multiple types of FIAs, our CFDR shows promising detection and recovery effectiveness. |
| title | Detection and Recovery Against Deep Neural Network Fault Injection Attacks Based on Contrastive Learning |
| topic | Machine Learning Artificial Intelligence Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.16766 |