Abstract
Recently, deep learning has emerged as a powerful technique for side-channel attacks, capable of even breaking common countermeasures. Still, trained models are generally large, and thus, performing evaluation becomes resource-intensive. The resource requirements increase in realistic settings where traces can be noisy, and countermeasures are active. In this work, we exploit mimicking to compress the learned models. We demonstrate up to 300 times compression of a state-of-the-art CNN. The mimic shallow network can also achieve much better accuracy as compared to when trained on original data and even reach the performance of a deeper network.
Original language | English |
---|---|
Title of host publication | 2020 57th ACM/IEEE Design Automation Conference, DAC 2020 |
Place of Publication | Piscataway |
Publisher | IEEE |
Pages | 1-6 |
Number of pages | 6 |
ISBN (Electronic) | 9781450367257 |
ISBN (Print) | 978-1-7281-5802-0 |
DOIs | |
Publication status | Published - 2020 |
Event | DAC 2020: 57th ACM/IEEE Design Automation Conference - San Francisco, United States Duration: 20 Jul 2020 → 24 Jul 2020 Conference number: 57th |
Publication series
Name | Proceedings - Design Automation Conference |
---|---|
Volume | 2020-July |
ISSN (Print) | 0738-100x |
Conference
Conference | DAC 2020 |
---|---|
Country/Territory | United States |
City | San Francisco |
Period | 20/07/20 → 24/07/20 |