Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking

In this paper we provide new theoretical and empirical evidences that gradient-based deep learning profiling attacks (DL-SCA) suffer from masking schemes. This occurs through an initial stall of the learning process: the so-called plateau effect. To understand why, we derive an analytical expressio...

Full description

Saved in:
Bibliographic Details
Main Authors: Nathan Rousselot, Karine Heydemann, Loïc Masure, Vincent Migairou
Format: Article
Language:English
Published: Ruhr-Universität Bochum 2025-06-01
Series:Transactions on Cryptographic Hardware and Embedded Systems
Subjects:
Online Access:https://tches.iacr.org/index.php/TCHES/article/view/12210
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849472240250257408
author Nathan Rousselot
Karine Heydemann
Loïc Masure
Vincent Migairou
author_facet Nathan Rousselot
Karine Heydemann
Loïc Masure
Vincent Migairou
author_sort Nathan Rousselot
collection DOAJ
description In this paper we provide new theoretical and empirical evidences that gradient-based deep learning profiling attacks (DL-SCA) suffer from masking schemes. This occurs through an initial stall of the learning process: the so-called plateau effect. To understand why, we derive an analytical expression of a DL-SCA model targeting simulated traces which enables us to study an analytical expression of the loss. By studying the loss landscape of this model, we show that not only do the magnitudes of the gradients decrease as the order of masking increases, but the loss landscape also exhibits a prominent saddle point interfering with the optimization process. From these observations, we (1) propose the usage of a second-order optimization algorithm mitigating the impact of low-gradient areas. In addition, we show how to leverage the intrinsic sparsity of valuable information in SCA traces to better pose the DL-SCA problem. To do so, we (2) propose to use the implicit regularization properties of the sparse mirror descent. These propositions are gathered in a new publicly available optimization algorithm, Scoop. Scoop combines second-order derivative of the loss function in the optimization process, with a sparse stochastic mirror descent. We experimentally show that Scoop pushes further the current limitations of DL-SCA against simulated traces, and outperforms the state-of-theart on the ASCADv1 dataset in terms of number of traces required to retrieve the key, perceived information and plateau length. Scoop also performs the first nonworst- case attack on the ASCADv2 dataset. On simulated traces, we show that using Scoop reduces the DL-SCA time complexity by the equivalent of one masking order.
format Article
id doaj-art-3e620ea364fb4303aa58f0c1e1587f7c
institution Kabale University
issn 2569-2925
language English
publishDate 2025-06-01
publisher Ruhr-Universität Bochum
record_format Article
series Transactions on Cryptographic Hardware and Embedded Systems
spelling doaj-art-3e620ea364fb4303aa58f0c1e1587f7c2025-08-20T03:24:35ZengRuhr-Universität BochumTransactions on Cryptographic Hardware and Embedded Systems2569-29252025-06-012025310.46586/tches.v2025.i3.56-80Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order MaskingNathan Rousselot0Karine Heydemann1Loïc Masure2Vincent Migairou3Thales, France; LIRMM, Univ. Montpellier, CNRS, FranceThales, FranceLIRMM, Univ. Montpellier, CNRS, FranceThales, France In this paper we provide new theoretical and empirical evidences that gradient-based deep learning profiling attacks (DL-SCA) suffer from masking schemes. This occurs through an initial stall of the learning process: the so-called plateau effect. To understand why, we derive an analytical expression of a DL-SCA model targeting simulated traces which enables us to study an analytical expression of the loss. By studying the loss landscape of this model, we show that not only do the magnitudes of the gradients decrease as the order of masking increases, but the loss landscape also exhibits a prominent saddle point interfering with the optimization process. From these observations, we (1) propose the usage of a second-order optimization algorithm mitigating the impact of low-gradient areas. In addition, we show how to leverage the intrinsic sparsity of valuable information in SCA traces to better pose the DL-SCA problem. To do so, we (2) propose to use the implicit regularization properties of the sparse mirror descent. These propositions are gathered in a new publicly available optimization algorithm, Scoop. Scoop combines second-order derivative of the loss function in the optimization process, with a sparse stochastic mirror descent. We experimentally show that Scoop pushes further the current limitations of DL-SCA against simulated traces, and outperforms the state-of-theart on the ASCADv1 dataset in terms of number of traces required to retrieve the key, perceived information and plateau length. Scoop also performs the first nonworst- case attack on the ASCADv2 dataset. On simulated traces, we show that using Scoop reduces the DL-SCA time complexity by the equivalent of one masking order. https://tches.iacr.org/index.php/TCHES/article/view/12210Side-channel AnalysisProfiling AttacksDeep learningMaskingOptimization
spellingShingle Nathan Rousselot
Karine Heydemann
Loïc Masure
Vincent Migairou
Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
Transactions on Cryptographic Hardware and Embedded Systems
Side-channel Analysis
Profiling Attacks
Deep learning
Masking
Optimization
title Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
title_full Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
title_fullStr Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
title_full_unstemmed Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
title_short Scoop: An Optimization Algorithm for Profiling Attacks against Higher-Order Masking
title_sort scoop an optimization algorithm for profiling attacks against higher order masking
topic Side-channel Analysis
Profiling Attacks
Deep learning
Masking
Optimization
url https://tches.iacr.org/index.php/TCHES/article/view/12210
work_keys_str_mv AT nathanrousselot scoopanoptimizationalgorithmforprofilingattacksagainsthigherordermasking
AT karineheydemann scoopanoptimizationalgorithmforprofilingattacksagainsthigherordermasking
AT loicmasure scoopanoptimizationalgorithmforprofilingattacksagainsthigherordermasking
AT vincentmigairou scoopanoptimizationalgorithmforprofilingattacksagainsthigherordermasking