Paper 2020/904

A Comparison of Weight Initializers in Deep Learning-based Side-channel Analysis

Huimin Li, Marina Krček, and Guilherme Perin


The usage of deep learning in profiled side-channel analysis requires a careful selection of neural network hyperparameters. In recent publications, different network architectures have been presented as efficient profiled methods against protected AES implementations. Indeed, completely different convolutional neural network models have presented similar performance against public side-channel traces databases. In this work, we analyze how weight initializers' choice influences deep neural networks' performance in the profiled side-channel analysis. Our results show that different weight initializers provide radically different behavior. We observe that even high-performing initializers can reach significantly different performance when conducting multiple training phases. Finally, we found that this hyperparameter is more dependent on the choice of dataset than other, commonly examined, hyperparameters. When evaluating the connections with other hyperparameters, the biggest connection is observed with activation functions.

Available format(s)
Publication info
Preprint. MINOR revision.
Weight initializationDeep learningSide-channel Analysis
Contact author(s)
h li-7 @ tudelft nl
m krcek @ tudelft nl
g perin @ tudelft nl
2020-08-19: revised
2020-07-18: received
See all versions
Short URL
Creative Commons Attribution


      author = {Huimin Li and Marina Krček and Guilherme Perin},
      title = {A Comparison of Weight Initializers in Deep Learning-based Side-channel Analysis},
      howpublished = {Cryptology ePrint Archive, Paper 2020/904},
      year = {2020},
      note = {\url{}},
      url = {}
Note: In order to protect the privacy of readers, does not use cookies or embedded third party content.