Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks
This work provides a critical review of the paper by Zaid et al. titled “Methodology for Efficient CNN Architectures in Profiling attacks”, which was published in TCHES Volume 2020, Issue 1. This work studies the design of CNN networks to perform side-channel analysis of multiple implementations of...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ruhr-Universität Bochum
2020-06-01
|
| Series: | Transactions on Cryptographic Hardware and Embedded Systems |
| Subjects: | |
| Online Access: | https://tches.iacr.org/index.php/TCHES/article/view/8586 |
| _version_ | 1818362447132622848 |
|---|---|
| author | Lennert Wouters Victor Arribas Benedikt Gierlichs Bart Preneel |
| author_facet | Lennert Wouters Victor Arribas Benedikt Gierlichs Bart Preneel |
| author_sort | Lennert Wouters |
| collection | DOAJ |
| description | This work provides a critical review of the paper by Zaid et al. titled “Methodology for Efficient CNN Architectures in Profiling attacks”, which was published in TCHES Volume 2020, Issue 1. This work studies the design of CNN networks to perform side-channel analysis of multiple implementations of the AES for embedded devices. Based on the authors’ code and public data sets, we were able to cross-check their results and perform a thorough analysis. We correct multiple misconceptions by carefully inspecting different elements of the model architectures proposed by Zaid et al. First, by providing a better understanding on the internal workings of these models, we can trivially reduce their number of parameters on average by 52%, while maintaining a similar performance. Second, we demonstrate that the convolutional filter’s size is not strictly related to the amount of misalignment in the traces. Third, we show that increasing the filter size and the number of convolutions actually improves the performance of a network. Our work demonstrates once again that reproducibility and review are important pillars of academic research. Therefore, we provide the reader with an online Python notebook which allows to reproduce some of our experiments1 and additional example code is made available on Github.2 |
| first_indexed | 2024-12-13T21:32:43Z |
| format | Article |
| id | doaj.art-fecbc54260a54268a373ddfb0f9e7cad |
| institution | Directory Open Access Journal |
| issn | 2569-2925 |
| language | English |
| last_indexed | 2024-12-13T21:32:43Z |
| publishDate | 2020-06-01 |
| publisher | Ruhr-Universität Bochum |
| record_format | Article |
| series | Transactions on Cryptographic Hardware and Embedded Systems |
| spelling | doaj.art-fecbc54260a54268a373ddfb0f9e7cad2022-12-21T23:30:46ZengRuhr-Universität BochumTransactions on Cryptographic Hardware and Embedded Systems2569-29252020-06-012020310.13154/tches.v2020.i3.147-168Revisiting a Methodology for Efficient CNN Architectures in Profiling AttacksLennert Wouters0Victor Arribas1Benedikt Gierlichs2Bart Preneel3imec-COSIC, KU Leuven Kasteelpark Arenberg 10, B-3001 Leuven-Heverlee, Belgiumimec-COSIC, KU Leuven Kasteelpark Arenberg 10, B-3001 Leuven-Heverlee, Belgiumimec-COSIC, KU Leuven Kasteelpark Arenberg 10, B-3001 Leuven-Heverlee, Belgiumimec-COSIC, KU Leuven Kasteelpark Arenberg 10, B-3001 Leuven-Heverlee, BelgiumThis work provides a critical review of the paper by Zaid et al. titled “Methodology for Efficient CNN Architectures in Profiling attacks”, which was published in TCHES Volume 2020, Issue 1. This work studies the design of CNN networks to perform side-channel analysis of multiple implementations of the AES for embedded devices. Based on the authors’ code and public data sets, we were able to cross-check their results and perform a thorough analysis. We correct multiple misconceptions by carefully inspecting different elements of the model architectures proposed by Zaid et al. First, by providing a better understanding on the internal workings of these models, we can trivially reduce their number of parameters on average by 52%, while maintaining a similar performance. Second, we demonstrate that the convolutional filter’s size is not strictly related to the amount of misalignment in the traces. Third, we show that increasing the filter size and the number of convolutions actually improves the performance of a network. Our work demonstrates once again that reproducibility and review are important pillars of academic research. Therefore, we provide the reader with an online Python notebook which allows to reproduce some of our experiments1 and additional example code is made available on Github.2https://tches.iacr.org/index.php/TCHES/article/view/8586Side-Channel AnalysisMachine LearningDeep Learning |
| spellingShingle | Lennert Wouters Victor Arribas Benedikt Gierlichs Bart Preneel Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks Transactions on Cryptographic Hardware and Embedded Systems Side-Channel Analysis Machine Learning Deep Learning |
| title | Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks |
| title_full | Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks |
| title_fullStr | Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks |
| title_full_unstemmed | Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks |
| title_short | Revisiting a Methodology for Efficient CNN Architectures in Profiling Attacks |
| title_sort | revisiting a methodology for efficient cnn architectures in profiling attacks |
| topic | Side-Channel Analysis Machine Learning Deep Learning |
| url | https://tches.iacr.org/index.php/TCHES/article/view/8586 |
| work_keys_str_mv | AT lennertwouters revisitingamethodologyforefficientcnnarchitecturesinprofilingattacks AT victorarribas revisitingamethodologyforefficientcnnarchitecturesinprofilingattacks AT benediktgierlichs revisitingamethodologyforefficientcnnarchitecturesinprofilingattacks AT bartpreneel revisitingamethodologyforefficientcnnarchitecturesinprofilingattacks |