When Single Event Upset Meets Deep Neural Networks: Observations, Explorations, and Remedies

Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in security sensitive systems. However, security-sensitive systems demand not only high perception performance, but also design robustness under variou...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the ASP-DAC ... Asia and South Pacific Design Automation Conference s. 163 - 168
Hlavní autoři: Yan, Zheyu, Shi, Yiyu, Liao, Wang, Hashimoto, Masanori, Zhou, Xichuan, Zhuo, Cheng
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.01.2020
Témata:
Hardware
Neural networks
Robustness
Sensitivity
Single event upsets
Software
Task analysis
ISSN:2153-697X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Deep Neural Network has proved its potential in various perception tasks and hence become an appealing option for interpretation and data processing in security sensitive systems. However, security-sensitive systems demand not only high perception performance, but also design robustness under various circumstances. Unlike prior works that study network robustness from software level, we investigate from hardware perspective about the impact of Single Event Upset (SEU) induced parameter perturbation (SIPP) on neural networks. We systematically define the fault models of SEU and then provide the definition of sensitivity to SIPP as the robustness measure for the network. We are then able to analytically explore the weakness of a network and summarize the key findings for the impact of SIPP on different types of bits in a floating point parameter, layer-wise robustness within the same network and impact of network depth. Based on those findings, we propose two remedy solutions to protect DNNs from SIPPs, which can mitigate accuracy degradation from 28% to 0.27% for ResNet with merely 0.24-bit SRAM area overhead per parameter.
ISSN:2153-697X
DOI:10.1109/ASP-DAC47756.2020.9045134