Convolutional Neural Network Accelerators: From Basic Design Principles to Advanced Security Applications
暫譯: 卷積神經網絡加速器：從基本設計原則到進階安全應用
Halak, Basel
- 出版商: Springer
- 出版日期: 2026-04-07
- 售價: $5,870
- 貴賓價: 9.5 折 $5,576
- 語言: 英文
- 頁數: 235
- 裝訂: Hardcover - also called cloth, retail trade, or trade
- ISBN: 3032085136
- ISBN-13: 9783032085139
-
相關分類:
DeepLearning
海外代購書籍(需單獨結帳)
商品描述
This book provides comprehensive coverage of the state-of-the-art in Convolutional Neural Network (CNN) hardware accelerator design, security, and its applications in hardware security. The first part gives a foundational understanding of CNN architectures, emphasizing their computational demands and the necessity for specialized hardware solutions. It also proposes an emulation method with open-source code to mimic CNN hardware accelerator behavior. The second part presents security applications of CNN models, featuring a case study in Network-on-Chip security. It covers threat modeling, countermeasures, and the use of alternative machine learning models to CNNs. The third part explains security threats throughout the AI model production lifecycle, including software vulnerabilities and hardware risks, and explores techniques to enhance the robustness of CNN hardware accelerators, focusing on preventing hardware Trojan and backdoor attacks and analyzing the vulnerability levels of different CNN layers.
商品描述(中文翻譯)
本書全面涵蓋了卷積神經網絡（Convolutional Neural Network, CNN）硬體加速器設計、安防及其在硬體安全中的應用的最新進展。第一部分提供了對CNN架構的基礎理解，強調其計算需求以及專用硬體解決方案的必要性。它還提出了一種模擬方法，並提供開源代碼以模擬CNN硬體加速器的行為。第二部分介紹了CNN模型的安全應用，並以片上網路（Network-on-Chip）安全為案例研究。內容涵蓋威脅建模、對策以及使用替代機器學習模型來取代CNN。第三部分解釋了在AI模型生產生命周期中存在的安全威脅，包括軟體漏洞和硬體風險，並探討了增強CNN硬體加速器穩健性的技術，重點在於防止硬體木馬和後門攻擊，並分析不同CNN層的脆弱性水平。
作者簡介
Dr Basel Halak is an Associate Professor of Secure Electronics and the Director of the Cyber Security Academy with the University of Southampton. He is also a leading European Masters in Embedded Computing Systems (EMECS). Dr. Halak is a visiting scholar at the Technical University of Kaiserslautern, the Norwegian University of Science and Technology, and the Polytechnic di Torino. He previously served as a visiting professor at the Kazakh-British Technical University 2017. Dr. Halak's expertise spans Digital Systems Design, Hardware Security, and Applied Cryptography. and he has authored over 120 refereed conference and journal papers and seven books, including the first textbook on Physically Unclonable Functions and the first book on Hardware Supply Chain Security. Beyond academia, Dr. Halak has collaborated extensively with industry such as ARM, Arqit, Schneider Electric, and Ericsson. Dr. Halak is the recipient of the Industrial Fellowship from the Royal Academy of Engineering and the National Teaching Fellowship awarded by the Advance Higher Education (HE) Academy. He actively contributes to the global research community as a member of technical program committees for leading conferences such as HOST, IEEE DATE, IEEE DAC, IVSW, ICCCA, ICCCS, MTV and EWME. He is an Associate Editor of IEEE access and a Guest Editor of the IET circuit devices and system journal. As of July 2025, he supervised to completion of 18 PhD students and 7 postdoctoral scholars.
作者簡介(中文翻譯)
巴塞爾·哈拉克博士（Dr. Basel Halak）是南安普敦大學（University of Southampton）安全電子學的副教授及網路安全學院的主任。他也是歐洲嵌入式計算系統碩士課程（European Masters in Embedded Computing Systems, EMECS）的領導者。哈拉克博士曾擔任凱瑟斯勞滕科技大學（Technical University of Kaiserslautern）、挪威科技大學（Norwegian University of Science and Technology）及都靈理工大學（Polytechnic di Torino）的訪問學者。他在2017年曾擔任哈薩克-英國技術大學（Kazakh-British Technical University）的訪問教授。哈拉克博士的專業領域涵蓋數位系統設計、硬體安全及應用密碼學，並已發表超過120篇經過審核的會議及期刊論文，以及七本書籍，包括第一本關於物理不可複製函數（Physically Unclonable Functions）的教科書和第一本關於硬體供應鏈安全的書籍。除了學術界，哈拉克博士還與業界廣泛合作，包括ARM、Arqit、施耐德電氣（Schneider Electric）及愛立信（Ericsson）。他獲得了英國皇家工程院（Royal Academy of Engineering）的工業獎學金及高等教育促進學院（Advance Higher Education, HE）頒發的國家教學獎學金。他作為技術程序委員會成員，積極參與全球研究社群，並參與HOST、IEEE DATE、IEEE DAC、IVSW、ICCCA、ICCCS、MTV及EWME等領先會議的組織工作。他是IEEE Access的副編輯及IET Circuit Devices and Systems期刊的客座編輯。截至2025年7月，他已指導完成18位博士生及7位博士後研究員的研究。
