{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T10:24:39Z","timestamp":1781519079262,"version":"3.54.1"},"reference-count":60,"publisher":"International Association for Cryptologic Research","license":[{"start":{"date-parts":[[2024,7,9]],"date-time":"2024-07-09T00:00:00Z","timestamp":1720483200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IACR CiC"],"accepted":{"date-parts":[[2024,9,2]]},"abstract":"<jats:p>Profiling side-channel analysis has gained widespread acceptance in both academic and industrial realms due to its robust capacity to unveil protected secrets, even in the presence of countermeasures. To harness this capability, an adversary must access a clone of the target device to acquire profiling measurements, labeling them with leakage models. The challenge of finding an effective leakage model, especially for a protected dataset with a low signal-to-noise ratio or weak correlation between actual leakages and labels, often necessitates an intuitive engineering approach, as otherwise, the attack will not perform well.<\/jats:p>\n          <jats:p>In this paper, we introduce a deep learning approach with a flexible leakage model, referred to as the multi-bit model. Instead of trying to learn a pre-determined representation of the target intermediate data, we utilize the concept of the stochastic model to decompose the label into bits. Then, the deep learning model is used to classify each bit independently. This versatile multi-bit model can adjust to existing leakage models like the Hamming weight and Most Significant Bit while also possessing the flexibility to adapt to complex leakage scenarios. To further improve the attack efficiency, we extend the multi-bit model to profile all 16 subkey bytes simultaneously, which requires negligible computational effort. The experimental results show that the proposed methods can efficiently break all key bytes across four considered datasets while the conventional leakage models fail. Our work signifies a significant step forward in deep learning-based side-channel attacks, showcasing a high degree of flexibility and efficiency with the proposed leakage model. <\/jats:p>","DOI":"10.62056\/ay4c3txol7","type":"journal-article","created":{"date-parts":[[2024,10,7]],"date-time":"2024-10-07T15:13:33Z","timestamp":1728314013000},"update-policy":"https:\/\/doi.org\/10.62056\/adfjwm02dj","source":"Crossref","is-referenced-by-count":9,"title":["Leakage Model-flexible Deep Learning-based Side-channel Analysis"],"prefix":"10.62056","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7139-732X","authenticated-orcid":false,"given":"Lichao","family":"Wu","sequence":"first","affiliation":[{"name":"Technical University of Darmstadt","place":["Darmstadt, Germany"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5665-8472","authenticated-orcid":false,"given":"Azade","family":"Rezaeezade","sequence":"additional","affiliation":[{"name":"Delft University of Technology","place":["Delft, The Netherlands"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7910-1159","authenticated-orcid":false,"given":"Amir","family":"Ali-pour","sequence":"additional","affiliation":[{"name":"\u00c9cole de technologie sup\u00e9rieure","place":["Montreal, Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3799-7636","authenticated-orcid":false,"given":"Guilherme","family":"Perin","sequence":"additional","affiliation":[{"name":"Leiden University","place":["Leiden, The Netherlands"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7509-4337","authenticated-orcid":false,"given":"Stjepan","family":"Picek","sequence":"additional","affiliation":[{"name":"Radboud University","place":["Nijmegen, The Netherlands"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"48349","published-online":{"date-parts":[[2024,10,7]]},"reference":[{"key":"ref1:DBLP:journals\/jce\/BenadjilaPSCD20","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1007\/s13389-019-00220-8","article-title":"Deep learning for side-channel analysis and introduction to\n  ASCAD database","volume":"10","author":"Ryad Benadjila","year":"2020","journal-title":"J. Cryptographic Engineering"},{"key":"ref2:perin2022exploring","doi-asserted-by":"publisher","first-page":"828","DOI":"10.46586\/tches.v2022.i4.828-861","article-title":"Exploring Feature Selection Scenarios for Deep\n  Learning-based Side-channel Analysis","author":"Guilherme Perin","year":"2022","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref3:wu2022choose","doi-asserted-by":"publisher","DOI":"10.1109\/tetc.2022.3218372","article-title":"I choose you: Automated hyperparameter tuning for deep\n  learning-based side-channel analysis","author":"Lichao Wu","year":"2022","journal-title":"IEEE Transactions on Emerging Topics in Computing"},{"key":"ref4:Zaid_Bossuet_Habrard_Venelli_2019","doi-asserted-by":"publisher","first-page":"1","DOI":"10.13154\/tches.v2020.i1.1-36","article-title":"Methodology for Efficient CNN Architectures in Profiling\n  Attacks","volume":"2020","author":"Gabriel Zaid","year":"2019","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref5:timon2019non","doi-asserted-by":"publisher","first-page":"107","DOI":"10.46586\/tches.v2019.i2.107-131","article-title":"Non-profiled deep learning-based side-channel attacks with\n  sensitivity analysis","author":"Benjamin Timon","year":"2019","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref6:schindler2005stochastic","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1007\/11545262_3","article-title":"A stochastic model for differential side channel\n  cryptanalysis","author":"Werner Schindler","year":"2005"},{"key":"ref7:SIMPLE-Crypto2023","volume-title":"SMAesH: Technical Documentation","author":"SIMPLE-Crypto","year":"2023"},{"key":"ref8:chari2003template","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1007\/3-540-36400-5_3","article-title":"Template attacks","author":"Suresh Chari","year":"2003"},{"key":"ref9:Heuser2012","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1007\/978-3-642-27954-6_23","article-title":"A New Difference Method for Side-Channel Analysis with\n  High-Dimensional Leakage Models","author":"Annelie Heuser","year":"2012"},{"key":"ref10:standaert2009unified","doi-asserted-by":"publisher","first-page":"443","DOI":"10.1007\/978-3-642-01001-9_26","article-title":"A unified framework for the analysis of side-channel key\n  recovery attacks","author":"Fran\u00e7ois-Xavier Standaert","year":"2009"},{"key":"ref11:GVK330798693","isbn-type":"print","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-72244-1","volume-title":"Applied multivariate statistical analysis","author":"Richard Arnold Johnson","year":"2002","ISBN":"https:\/\/id.crossref.org\/isbn\/0130925535"},{"key":"ref12:DBLP:conf\/cardis\/ChoudaryK13","series-title":"LNCS","isbn-type":"print","doi-asserted-by":"publisher","first-page":"253","DOI":"10.1007\/978-3-319-08302-5_17","article-title":"Efficient Template Attacks","volume":"8419","author":"Omar Choudary","year":"2013","ISBN":"https:\/\/id.crossref.org\/isbn\/9783319083018"},{"key":"ref13:8836108","doi-asserted-by":"publisher","first-page":"2802","DOI":"10.1109\/tvlsi.2019.2937365","article-title":"A Systematic Evaluation of Profiling Through Focused Feature\n  Selection","volume":"27","author":"Stjepan Picek","year":"2019","journal-title":"IEEE Transactions on Very Large Scale Integration (VLSI)\n  Systems"},{"key":"ref14:wold1987principal","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1016\/0169-7439(87)80084-9","article-title":"Principal component analysis","volume":"2","author":"Svante Wold","year":"1987","journal-title":"Chemometrics and intelligent laboratory systems"},{"key":"ref15:archambeau2006template","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/11894063_1","article-title":"Template attacks in principal subspaces","author":"C\u00e9dric Archambeau","year":"2006"},{"key":"ref16:batina2012getting","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1007\/978-3-642-27954-6_24","article-title":"Getting more from PCA: first results of using principal\n  component analysis for extensive power analysis","author":"Lejla Batina","year":"2012"},{"key":"ref17:standaert2008using","doi-asserted-by":"publisher","first-page":"411","DOI":"10.1007\/978-3-540-85053-3_26","article-title":"Using subspace-based template attacks to compare and combine\n  power and electromagnetic information leakages","author":"Fran\u00e7ois-Xavier Standaert","year":"2008"},{"key":"ref18:gierlichs2006templates","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1007\/11894063_2","article-title":"Templates vs. stochastic methods","author":"Benedikt Gierlichs","year":"2006"},{"key":"ref19:cassiers2023efficient","doi-asserted-by":"publisher","first-page":"270","DOI":"10.46586\/tches.v2023.i3.270-293","article-title":"Efficient regression-based linear discriminant analysis for\n  side-channel security evaluations: Towards analytical attacks against 32-bit\n  implementations","volume":"2023","author":"Ga\u00ebtan Cassiers","year":"2023","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref20:LL:CARDIS13","series-title":"Lecture Notes in Computer Science","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-14123-7_5","article-title":"A Machine Learning Approach Against a Masked AES","author":"Liran Lerman","year":"2013"},{"key":"ref21:DBLP:journals\/jce\/HospodarGMVV11","doi-asserted-by":"publisher","first-page":"293","DOI":"10.1007\/s13389-011-0023-x","article-title":"Machine learning in side-channel analysis: a first study","volume":"1","author":"Gabriel Hospodar","year":"2011","journal-title":"J. Cryptogr. Eng."},{"key":"ref22:Picek2017","doi-asserted-by":"publisher","first-page":"343","DOI":"10.1007\/s13389-017-0172-7","article-title":"Template attack versus Bayes classifier","volume":"7","author":"Stjepan Picek","year":"2017","journal-title":"Journal of Cryptographic Engineering"},{"key":"ref23:maghrebi2016breaking","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/978-3-319-49445-6_1","article-title":"Breaking cryptographic implementations using deep learning\n  techniques","author":"Houssem Maghrebi","year":"2016"},{"key":"ref24:cagli2017convolutional","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1007\/978-3-319-66787-4_3","article-title":"Convolutional neural networks with data augmentation against\n  jitter-based countermeasures: Profiling attacks without pre-processing","author":"Eleonora Cagli","year":"2017"},{"key":"ref25:kim2019make","doi-asserted-by":"publisher","first-page":"148","DOI":"10.46586\/tches.v2019.i3.148-179","article-title":"Make some noise. unleashing the power of convolutional\n  neural networks for profiled side-channel analysis","author":"Jaehun Kim","year":"2019","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref26:Rezaeezade2022Regularization","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s13389-024-00361-5","article-title":"Regularizers to the rescue: fighting overfitting in deep\n  learning-based side-channel analysis","author":"Azade Rezaeezade","year":"2024","journal-title":"Journal of Cryptographic Engineering"},{"key":"ref27:Hern2018augment","article-title":"Data augmentation instead of explicit regularization","volume":"abs\/1806.03852","author":"Alex Hern\u00e1ndez-Garc\u00eda","year":"2018","journal-title":"CoRR"},{"key":"ref28:Wouters_Arribas_Gierlichs_Preneel_2020","doi-asserted-by":"publisher","first-page":"147","DOI":"10.13154\/tches.v2020.i3.147-168","article-title":"Revisiting a Methodology for Efficient CNN Architectures in\n  Profiling Attacks","volume":"2020","author":"Lennert Wouters","year":"2020","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref29:wu2023label","doi-asserted-by":"publisher","DOI":"10.1109\/tifs.2023.3287728","article-title":"Label Correlation in Deep Learning-based Side-channel\n  Analysis","author":"Lichao Wu","year":"2023","journal-title":"IEEE Transactions on Information Forensics and Security"},{"key":"ref30:zaid2023conditional","doi-asserted-by":"publisher","first-page":"310","DOI":"10.46586\/tches.v2023.i2.310-357","article-title":"Conditional variational autoencoder based on stochastic\n  attacks","author":"Gabriel Zaid","year":"2023","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref31:zhang2020multilabel","doi-asserted-by":"publisher","first-page":"1207","DOI":"10.1109\/TCAD.2020.3033495","article-title":"Multilabel deep learning-based side-channel attack","volume":"40","author":"Libang Zhang","year":"2020","journal-title":"IEEE Transactions on Computer-Aided Design of Integrated\n  Circuits and Systems"},{"key":"ref32:10.1145\/3569577","doi-asserted-by":"publisher","DOI":"10.1145\/3569577","article-title":"SoK: Deep Learning-Based Physical Side-Channel Analysis","volume":"55","author":"Stjepan Picek","year":"2023","journal-title":"ACM Comput. Surv.","ISSN":"https:\/\/id.crossref.org\/issn\/0360-0300","issn-type":"electronic"},{"key":"ref33:choudary2015efficient","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1007\/978-3-319-16763-3_6","article-title":"Efficient stochastic methods: Profiled attacks beyond 8\n  bits","author":"Marios O Choudary","year":"2015"},{"key":"ref34:standaert2009compare","doi-asserted-by":"publisher","first-page":"485","DOI":"10.1007\/978-3-642-01957-9_30","article-title":"How to compare profiled side-channel attacks?","author":"Fran\u00e7ois-Xavier Standaert","year":"2009"},{"key":"ref35:gierlichs2008mutual","doi-asserted-by":"publisher","first-page":"426","DOI":"10.1007\/978-3-540-85053-3_27","article-title":"Mutual information analysis","author":"Benedikt Gierlichs","year":"2008"},{"key":"ref36:bronchain2019leakage","doi-asserted-by":"publisher","first-page":"713","DOI":"10.1007\/978-3-030-26948-7_25","article-title":"Leakage certification revisited: Bounding model errors in\n  side-channel security evaluations","author":"Olivier Bronchain","year":"2019"},{"key":"ref37:zhang2020novel","doi-asserted-by":"publisher","first-page":"73","DOI":"10.46586\/tches.v2020.i3.73-96","article-title":"A novel evaluation metric for deep learning-based side\n  channel analysis and its extended application to imbalanced data","author":"Jiajia Zhang","year":"2020","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref38:Picek_Heuser_Jovic_Bhasin_Regazzoni_2018","doi-asserted-by":"publisher","first-page":"209","DOI":"10.13154\/tches.v2019.i1.209-237","article-title":"The Curse of Class Imbalance and Conflicting Metrics with\n  Machine Learning for Side-channel Evaluations","volume":"2019","author":"Stjepan Picek","year":"2018","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref39:caruana1997multitask","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1007\/978-1-4615-5529-2_5","article-title":"Multitask learning","volume":"28","author":"Rich Caruana","year":"1997","journal-title":"Machine learning"},{"key":"ref40:ruder2017overview","article-title":"An overview of multi-task learning in deep neural networks","author":"Sebastian Ruder","year":"2017","journal-title":"arXiv preprint arXiv:1706.05098"},{"key":"ref41:zhang2021survey","doi-asserted-by":"publisher","first-page":"5586","DOI":"10.1109\/TKDE.2021.3070203","article-title":"A survey on multi-task learning","volume":"34","author":"Yu Zhang","year":"2021","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"ref42:maghrebi2020deep","article-title":"Deep learning based side-channel attack: a new profiling\n  methodology based on multi-label classification","author":"Houssem Maghrebi","year":"2020","journal-title":"Cryptology ePrint Archive"},{"key":"ref43:picek2018performance","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1007\/978-3-030-05072-6_10","article-title":"On the performance of convolutional neural networks for\n  side-channel analysis","author":"Stjepan Picek","year":"2018"},{"key":"ref44:won2020push","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1007\/978-3-030-65299-9_18","article-title":"Push for more: On comparison of data augmentation and SMOTE\n  with optimised deep learning architecture for side-channel","author":"Yoo-Seung Won","year":"2020"},{"key":"ref45:ho2002simple","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1023\/a:1021251113462","article-title":"Simple explanation of the no-free-lunch theorem and its\n  implications","volume":"115","author":"Yu-Chi Ho","year":"2002","journal-title":"Journal of optimization theory and applications"},{"key":"ref46:klambauer2017self","first-page":"971","article-title":"Self-normalizing neural networks","volume-title":"Advances in Neural Information Processing Systems","author":"G\u00fcnter Klambauer","year":"2017"},{"key":"ref47:bridle1990probabilistic","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1007\/978-3-642-76153-9_28","article-title":"Probabilistic interpretation of feedforward classification\n  network outputs, with relationships to statistical pattern recognition","author":"John S Bridle","year":"1990"},{"key":"ref48:vasselle2023spatial","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s13389-022-00307-9","article-title":"Spatial dependency analysis to extract information from\n  side-channel mixtures: extended version","author":"Aur\u00e9lien Vasselle","year":"2023","journal-title":"Journal of Cryptographic Engineering"},{"key":"ref49:gohr2021subsampling","doi-asserted-by":"publisher","first-page":"567","DOI":"10.1007\/978-3-030-81652-0_22","article-title":"Subsampling and knowledge distillation on adversarial\n  examples: New techniques for deep learning based side channel evaluations","author":"Aron Gohr","year":"2021"},{"key":"ref50:DBLP:journals\/tc\/CassiersGLS21","doi-asserted-by":"publisher","first-page":"1677","DOI":"10.1109\/TC.2020.3022979","article-title":"Hardware Private Circuits: From Trivial Composition to Full\n  Verification","volume":"70","author":"Ga\u00ebtan Cassiers","year":"2021","journal-title":"IEEE Trans. Computers"},{"key":"ref51:DBLP:conf\/asiacrypt\/Veyrat-CharvillonGS14","series-title":"Lecture Notes in Computer Science","doi-asserted-by":"publisher","first-page":"282","DOI":"10.1007\/978-3-662-45611-8_15","article-title":"Soft Analytical Side-Channel Attacks","volume":"8873","author":"Nicolas Veyrat-Charvillon","year":"2014"},{"key":"ref52:smaesh_leaderboard","volume-title":"SMAesH Challenge Leaderboard","author":"Simple Crypto","year":"2024"},{"key":"ref53:keskar2016large","article-title":"On large-batch training for deep learning: Generalization\n  gap and sharp minima","author":"Nitish Shirish Keskar","year":"2016","journal-title":"arXiv preprint arXiv:1609.04836"},{"key":"ref54:masters2018revisiting","article-title":"Revisiting small batch training for deep neural networks","author":"Dominic Masters","year":"2018","journal-title":"arXiv preprint arXiv:1804.07612"},{"key":"ref55:Rijsdijk_Wu_Perin_Picek_2021","doi-asserted-by":"publisher","first-page":"677","DOI":"10.46586\/tches.v2021.i3.677-707","article-title":"Reinforcement Learning for Hyperparameter Tuning in Deep\n  Learning-based Side-channel Analysis","volume":"2021","author":"Jorai Rijsdijk","year":"2021","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref56:acharya2022information","doi-asserted-by":"publisher","DOI":"10.46586\/tches.v2023.i1.401-437","article-title":"Information theory-based evolution of neural networks for\n  side-channel analysis","author":"Rabin Y Acharya","year":"2022","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref57:gohr2022breaking","doi-asserted-by":"publisher","first-page":"397","DOI":"10.46586\/tches.v2022.i4.397-437","article-title":"Breaking masked implementations of the clyde-cipher by means\n  of side-channel analysis: A report on the ches challenge side-channel contest\n  2020","author":"Aron Gohr","year":"2022","journal-title":"IACR Transactions on Cryptographic Hardware and Embedded\n  Systems"},{"key":"ref58:fukuda2023profiling","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1587\/transfun.2022cip0015","article-title":"Profiling Deep Learning Side-Channel Attacks Using\n  Multi-Label against AES Circuits with RSM Countermeasure","volume":"106","author":"Yuta Fukuda","year":"2023","journal-title":"IEICE Transactions on Fundamentals of Electronics,\n  Communications and Computer Sciences"},{"key":"ref59:hamalainen2006design","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1109\/dsd.2006.40","article-title":"Design and implementation of low-area and low-power AES\n  encryption hardware core","author":"Panu Hamalainen","year":"2006"},{"key":"ref60:DBLP:conf\/cardis\/2013","series-title":"LNCS","isbn-type":"print","volume-title":"CARDIS","volume":"8419","year":"2014","ISBN":"https:\/\/id.crossref.org\/isbn\/9783319083018"}],"container-title":["IACR Communications in Cryptology"],"original-title":[],"language":"en","deposited":{"date-parts":[[2024,12,10]],"date-time":"2024-12-10T21:28:42Z","timestamp":1733866122000},"score":1,"resource":{"primary":{"URL":"https:\/\/cic.iacr.org\/p\/1\/3\/41"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,7]]},"references-count":60,"URL":"https:\/\/doi.org\/10.62056\/ay4c3txol7","archive":["Internet Archive","Internet Archive"],"relation":{},"ISSN":["3006-5496"],"issn-type":[{"value":"3006-5496","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,7]]},"assertion":[{"value":"2024-07-09","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-09-02","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"cc1-3-116"}}