Paper 2019/1477 Original: eprint.iacr.org/2019/1477 ↗
Kilroy was here: The First Step Towards Explainability of Neural Networks in Profiled Side-channel Analysis
Abstract
While several works have explored the application of deep learning for efficient profiled side-channel analysis, explainability or in other words what neural networks learn remains a rather untouched topic. As a first step, this paper explores the Singular Vector Canonical Correlation Analysis (SVCCA) tool to interpret what neural networks learn while training on different side-channel datasets, by concentrating on deep layers of the network. Information from SVCCA can help, to an extent, with several practical problems in a profiled side-channel analysis like portability issue and criteria to choose a number of layers/neurons to fight portability, provide insight on the correct size of training dataset and detect deceptive conditions like over-specialization of networks.
Metadata
- Available format(s)
PDF- Publication info
- Preprint.
- History
- 2019-12-23: last updated
- 2019-12-23: published
- Short URL
- https://ia.cr/2019/1477
- License
- https://creativecommons.org/licenses/by/4.0/
BibTeX
@misc{cryptoeprint:2019/1477,
author = {Daan van der Valk and Stjepan Picek and Shivam Bhasin},
title = {Kilroy was here: The First Step Towards Explainability of Neural Networks in Profiled Side-channel Analysis},
howpublished = {Cryptology {ePrint} Archive, Paper 2019/1477},
year = {2019},
url = {https://eprint.iacr.org/2019/1477}
}