This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
CPSGD: A Novel Optimization Algorithm and Its Application in Side-Channel Analysis
by
Yifan Zhang
Yifan Zhang 1,
Di Zhao
Di Zhao
Dr. Di Zhao has a position as a professor and is currently working at the School of Mathematical He [...]
Dr. Di Zhao has a position as a professor and is currently working at the School of Mathematical Sciences, Beihang University. He received his master’s degree from Fudan University in 1993 and his doctorate from the Institute of Mathematics, Chinese Academy of Sciences in 1999. He was a visiting scholar at the Hong Kong University of Science and Technology from 1999 to 2000, a postdoctoral fellow at the Institute of Applied Mathematics, Chinese Academy of Sciences from 2000 to 2001, and an associate professor and professor/doctoral supervisor at the Beihang University from 2001 to date. His research areas are mainly information security, matrix theory application and computing, signal processing, artificial intelligence, and software development. His main courses include matrix theory, advanced algebra, and comprehensive courses for doctoral students. He has published more than 70 academic papers. He serves as a project review expert for the Beijing Natural Science Foundation, a review expert for the Ministry of Education’s Degree Center, and a doctoral dissertation sampling expert.
1,2,
Hongyi Li
Hongyi Li
Prof. Hongyi Li was a postgraduate student at Renmin University of China (1999) and a doctoral at a [...]
Prof. Hongyi Li was a postgraduate student at Renmin University of China (1999) and a doctoral student at Beihang University (2006). She was a visiting scholar at York University in Canada and the University of Macau (2010–2014). Currently, she has a position as a professor and doctoral supervisor and is working at the School of Mathematical Sciences, Beihang University. Her research areas are information security based on artificial intelligence, signal processing, software development, big data algorithm analysis and processing, applied mathematics, and numerical algebra. She is a project review expert for the National Natural Science Foundation of China and the Beijing Natural Science Foundation of China, an expert reviewer at the Degree Center of the Ministry of Education and an expert in sampling doctoral dissertations, and a mentor for the “Talent Program” jointly implemented by the China Association for Science and Technology and the Ministry of Education. She has presided over many scientific research projects, including the National Natural Science Foundation, National Key Civil Aerospace sub-topics, and National 973 Project sub-topics.
1,2,* and
Chengwei Pan
Chengwei Pan
Chengwei Pan is currently an associate professor at the Institute of Artificial Intelligence of He [...]
Chengwei Pan is currently an associate professor at the Institute of Artificial Intelligence of Beihang University. He received the B.S. degree from Beihang University in 2012 and the Ph.D. degree from the Department of Computer Science at Peking University in 2018. His research interests include virtual/augmented reality, computer graphics, computer vision, and medical image processing, etc.
3,4,*
1
School of Cyber Science and Technology, Beihang University, Beijing 100191, China
2
School of Mathematical Sciences, Beihang University, Beijing 100191, China
3
Institute of Artificial Intelligence, Beihang University, Beijing 100191, China
4
Key Laboratory of Mathematics, Informatics and Behavioral Semantics, Ministry of Education, Beijing 100191, China
*
Authors to whom correspondence should be addressed.
Mathematics 2024, 12(15), 2355; https://doi.org/10.3390/math12152355 (registering DOI)
Submission received: 2 June 2024
/
Revised: 21 July 2024
/
Accepted: 25 July 2024
/
Published: 28 July 2024
Abstract
In recent years, side-channel analysis based on deep learning has garnered significant attention from researchers. A pivotal reason for this lies in the fact that deep learning-based side-channel analysis requires minimal preprocessing of side-channel data. The automatic feature extraction property of deep learning methods drastically reduces the workload for researchers, enabling them to focus more on the core issues of side-channel analysis, namely, extracting sensitive information by attacking devices. However, in prior studies, most scholars have concentrated more on the model construction process, with little research focusing on the choice of optimizers.This paper explores a novel deep learning-based optimization algorithm—CPSGD (combined projection stochastic gradient descent). The algorithm comprises two variants, designed, respectively, for unprotected side-channel analysis (CPSGD1) and desynchronized side-channel analysis (CPSGD2), and their convergence has been theoretically proven. Experimental results demonstrate that, while maintaining the neural network structure unchanged, CPSGD1 exhibits the best performance on unprotected datasets compared to other publicly available optimizers, whereas CPSGD2 performs optimally on desynchronized datasets.
Share and Cite
MDPI and ACS Style
Zhang, Y.; Zhao, D.; Li, H.; Pan, C.
CPSGD: A Novel Optimization Algorithm and Its Application in Side-Channel Analysis. Mathematics 2024, 12, 2355.
https://doi.org/10.3390/math12152355
AMA Style
Zhang Y, Zhao D, Li H, Pan C.
CPSGD: A Novel Optimization Algorithm and Its Application in Side-Channel Analysis. Mathematics. 2024; 12(15):2355.
https://doi.org/10.3390/math12152355
Chicago/Turabian Style
Zhang, Yifan, Di Zhao, Hongyi Li, and Chengwei Pan.
2024. "CPSGD: A Novel Optimization Algorithm and Its Application in Side-Channel Analysis" Mathematics 12, no. 15: 2355.
https://doi.org/10.3390/math12152355
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.