TY - JOUR
T1 - Reinforcement learning for hyperparameter tuning in deep learning-based side-channel analysis
AU - Rijsdijk, Jorai
AU - Wu, Lichao
AU - Perin, Guilherme
AU - Picek, Stjepan
PY - 2021
Y1 - 2021
N2 - Deep learning represents a powerful set of techniques for profiling side-channel analysis. The results in the last few years show that neural network architectures like multilayer perceptron and convolutional neural networks give strong attack performance where it is possible to break targets protected with various coun-termeasures. Considering that deep learning techniques commonly have a plethora of hyperparameters to tune, it is clear that such top attack results can come with a high price in preparing the attack. This is especially problematic as the side-channel community commonly uses random search or grid search techniques to look for the best hyperparameters. In this paper, we propose to use reinforcement learning to tune the convolutional neural network hyperparameters. In our framework, we investigate the Q-Learning paradigm and develop two reward functions that use side-channel metrics. We mount an investigation on three commonly used datasets and two leakage models where the results show that reinforcement learning can find convolutional neural networks exhibiting top performance while having small numbers of trainable parameters. We note that our approach is automated and can be easily adapted to different datasets. Several of our newly developed architectures outperform the current state-of-the-art results. Finally, we make our source code publicly available.1.
AB - Deep learning represents a powerful set of techniques for profiling side-channel analysis. The results in the last few years show that neural network architectures like multilayer perceptron and convolutional neural networks give strong attack performance where it is possible to break targets protected with various coun-termeasures. Considering that deep learning techniques commonly have a plethora of hyperparameters to tune, it is clear that such top attack results can come with a high price in preparing the attack. This is especially problematic as the side-channel community commonly uses random search or grid search techniques to look for the best hyperparameters. In this paper, we propose to use reinforcement learning to tune the convolutional neural network hyperparameters. In our framework, we investigate the Q-Learning paradigm and develop two reward functions that use side-channel metrics. We mount an investigation on three commonly used datasets and two leakage models where the results show that reinforcement learning can find convolutional neural networks exhibiting top performance while having small numbers of trainable parameters. We note that our approach is automated and can be easily adapted to different datasets. Several of our newly developed architectures outperform the current state-of-the-art results. Finally, we make our source code publicly available.1.
KW - Convolutional Neural Networks
KW - Deep Learning
KW - Hyperparameter tuning
KW - Q-policy
KW - Reinforcement Learning
KW - Reward
KW - Side-channel Analysis
UR - http://www.scopus.com/inward/record.url?scp=85118420291&partnerID=8YFLogxK
U2 - 10.46586/tches.v2021.i3.677-707
DO - 10.46586/tches.v2021.i3.677-707
M3 - Article
AN - SCOPUS:85118420291
SN - 2569-2925
VL - 2021
SP - 677
EP - 707
JO - IACR Transactions on Cryptographic Hardware and Embedded Systems
JF - IACR Transactions on Cryptographic Hardware and Embedded Systems
IS - 3
ER -