Optimal Dimensionality Reduction using Conditional Variational AutoEncoder

The benefits of using Deep Learning techniques to enhance side-channel attacks performances have been demonstrated over recent years. Most of the work carried out since then focuses on discriminative models. However, one of their major limitations is the lack of theoretical results. Indeed, this la...

Full description

Saved in:
Bibliographic Details
Main Authors: Sana Boussam, Mathieu Carbone, Benoît Gérard, Guénaël Renault, Gabriel Zaid
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/12214
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850233285162041344
author Sana Boussam
Mathieu Carbone
Benoît Gérard
Guénaël Renault
Gabriel Zaid
author_facet Sana Boussam
Mathieu Carbone
Benoît Gérard
Guénaël Renault
Gabriel Zaid
author_sort Sana Boussam
collection DOAJ
description The benefits of using Deep Learning techniques to enhance side-channel attacks performances have been demonstrated over recent years. Most of the work carried out since then focuses on discriminative models. However, one of their major limitations is the lack of theoretical results. Indeed, this lack of theoretical results, especially concerning the choice of neural network architecture to consider or the loss to prioritize to build an optimal model, can be problematic for both attackers and evaluators. Recently, Zaid et al. addressed this problem by proposing a generative model that bridges conventional profiled attacks and deep learning techniques, thus providing a model that is both explicable and interpretable. Nevertheless the proposed model has several limitations. Indeed, the architecture is too complex, higher-order attacks cannot be mounted and desynchronization is not handled by this model. In this paper, we address the first limitation namely the architecture complexity, as without a simpler model, the other limitations cannot be treated properly. To do so, we propose a new generative model that relies on solid theoretical results. This model is based on conditional variational autoencoder and converges towards the optimal statistical model i.e. it performs an optimal attack. By building on and extending the state-of-the-art theoretical works on dimensionality reduction, we integrate into this neural network an optimal dimensionality reduction i.e. a dimensionality reduction that is achieved without any loss of information. This results in a gain of O(D), with D the dimension of traces, compared to Zaid et al. neural network in terms of architecture complexity, while at the same time enhancing the explainability and interpretability. In addition, we propose a new attack strategy based on our neural network, which reduces the attack complexity of generative models from O(N) to O(1), with N the number of generated traces. We validate all our theoretical results experimentally using extensive simulations and various publicly available datasets covering symmetric, asymmetric pre and post-quantum cryptography implementations.
format Article
id doaj-art-95892bfa2003433ea20557dbfd1de1dd
institution OA Journals
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-95892bfa2003433ea20557dbfd1de1dd2025-08-20T02:02:57ZengRuhr-Universität BochumTransactions on Cryptographic Hardware and Embedded Systems2569-29252025-06-012025310.46586/tches.v2025.i3.164-211Optimal Dimensionality Reduction using Conditional Variational AutoEncoderSana Boussam0Mathieu Carbone1Benoît Gérard2Guénaël Renault3Gabriel Zaid4LIX, INRIA, CNRS, École Polytechnique, Institut Polytechnique de Paris, Palaiseau, France; Thales ITSEF, Toulouse, FranceThales ITSEF, Toulouse, FranceANSSI, Paris, FranceANSSI, Paris, France; LIX, INRIA, CNRS, École Polytechnique, Institut Polytechnique de Paris, Palaiseau, FranceThales ITSEF, Toulouse, France The benefits of using Deep Learning techniques to enhance side-channel attacks performances have been demonstrated over recent years. Most of the work carried out since then focuses on discriminative models. However, one of their major limitations is the lack of theoretical results. Indeed, this lack of theoretical results, especially concerning the choice of neural network architecture to consider or the loss to prioritize to build an optimal model, can be problematic for both attackers and evaluators. Recently, Zaid et al. addressed this problem by proposing a generative model that bridges conventional profiled attacks and deep learning techniques, thus providing a model that is both explicable and interpretable. Nevertheless the proposed model has several limitations. Indeed, the architecture is too complex, higher-order attacks cannot be mounted and desynchronization is not handled by this model. In this paper, we address the first limitation namely the architecture complexity, as without a simpler model, the other limitations cannot be treated properly. To do so, we propose a new generative model that relies on solid theoretical results. This model is based on conditional variational autoencoder and converges towards the optimal statistical model i.e. it performs an optimal attack. By building on and extending the state-of-the-art theoretical works on dimensionality reduction, we integrate into this neural network an optimal dimensionality reduction i.e. a dimensionality reduction that is achieved without any loss of information. This results in a gain of O(D), with D the dimension of traces, compared to Zaid et al. neural network in terms of architecture complexity, while at the same time enhancing the explainability and interpretability. In addition, we propose a new attack strategy based on our neural network, which reduces the attack complexity of generative models from O(N) to O(1), with N the number of generated traces. We validate all our theoretical results experimentally using extensive simulations and various publicly available datasets covering symmetric, asymmetric pre and post-quantum cryptography implementations. https://tches.iacr.org/index.php/TCHES/article/view/12214Side-Channel AttacksDeep LearningOptimal Dimensionality ReductionOptimal AttacksGenerative ModelsVariational AutoEncoder
spellingShingle Sana Boussam
Mathieu Carbone
Benoît Gérard
Guénaël Renault
Gabriel Zaid
Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
Transactions on Cryptographic Hardware and Embedded Systems
Side-Channel Attacks
Deep Learning
Optimal Dimensionality Reduction
Optimal Attacks
Generative Models
Variational AutoEncoder
title Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
title_full Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
title_fullStr Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
title_full_unstemmed Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
title_short Optimal Dimensionality Reduction using Conditional Variational AutoEncoder
title_sort optimal dimensionality reduction using conditional variational autoencoder
topic Side-Channel Attacks
Deep Learning
Optimal Dimensionality Reduction
Optimal Attacks
Generative Models
Variational AutoEncoder
url https://tches.iacr.org/index.php/TCHES/article/view/12214
work_keys_str_mv AT sanaboussam optimaldimensionalityreductionusingconditionalvariationalautoencoder
AT mathieucarbone optimaldimensionalityreductionusingconditionalvariationalautoencoder
AT benoitgerard optimaldimensionalityreductionusingconditionalvariationalautoencoder
AT guenaelrenault optimaldimensionalityreductionusingconditionalvariationalautoencoder
AT gabrielzaid optimaldimensionalityreductionusingconditionalvariationalautoencoder