Paper 2022/1507

Label Correlation in Deep Learning-based Side-channel Analysis

Lichao Wu, Delft University of Technology
Léo Weissbart, Radboud University Nijmegen, Delft University of Technology
Marina Krček, Delft University of Technology
Huimin Li, Delft University of Technology
Guilherme Perin, Radboud University Nijmegen
Lejla Batina, Radboud University Nijmegen
Stjepan Picek, Radboud University Nijmegen

The efficiency of the profiling side-channel analysis can be significantly improved with machine learning techniques. Although powerful, a fundamental machine learning limitation of being data-hungry received little attention in the side-channel community. In practice, the maximum number of leakage traces that evaluators/attackers can obtain is constrained by the scheme requirements or the limited accessibility of the target. Even worse, various countermeasures in modern devices increase the conditions on the profiling size to break the target. This work demonstrates a practical approach to dealing with the lack of profiling traces. Instead of learning from a one-hot encoded label, transferring the labels to their distribution can significantly speed up the convergence of guessing entropy. By studying the relationship between all possible key candidates, we propose a new metric, denoted Label Correlation (LC), to evaluate the generalization ability of the profiling model. We validate LC with two common use cases: early stopping and network architecture search, and the results indicate its superior performance.

Available format(s)
Publication info
Published elsewhere. IEEE Transactions on Information Forensics and Security
Side-channel AnalysisProfiling AnalysisDeep LearningLabel DistributionProfiling Model Fitting
Contact author(s)
lichao wu9 @ gmail com
l weissbart @ cs ru nl
m krcek @ tudelft nl
h li-7 @ tudelft nl
guilhermeperin7 @ gmail com
lejla @ cs ru nl
picek stjepan @ gmail com
2023-05-30: revised
2022-11-01: received
See all versions
Short URL
Creative Commons Attribution


      author = {Lichao Wu and Léo Weissbart and Marina Krček and Huimin Li and Guilherme Perin and Lejla Batina and Stjepan Picek},
      title = {Label Correlation in Deep Learning-based Side-channel Analysis},
      howpublished = {Cryptology ePrint Archive, Paper 2022/1507},
      year = {2022},
      note = {\url{}},
      url = {}
Note: In order to protect the privacy of readers, does not use cookies or embedded third party content.