Title Number systems for deep neural network architectures / Ghada Alsuhli, Vasilis Sakellariou, Hani Saleh, Mahmoud Al-Qutayri, Baker Mohammad, Thanos Stouraitis

Published Cham : Springer, [2024]

Click on the following:
Springer eBooks

Copies

Description 1 online resource (xi, 94 pages) : illustrations (some color) Series Synthesis lectures on engineering, science, and technology, 2690-0327 Synthesis Lectures on Engineering, Science, and Technology (Springer (Firm)), 2690-0327 Contents Introduction -- Conventional number systems -- DNN architectures based on Logarithmic Number System (LNS) -- DNN architectures based on Residue Number System (RNS) -- DNN architectures based on Block Floating Point (BFP) number system -- DNN architectures based on Dynamic Fixed Point (DFXP) number system -- DNN architectures based on Posit number system Summary This book provides readers a comprehensive introduction to alternative number systems for more efficient representations of Deep Neural Network (DNN) data. Various number systems (conventional/unconventional) exploited for DNNs are discussed, including Floating Point (FP), Fixed Point (FXP), Logarithmic Number System (LNS), Residue Number System (RNS), Block Floating Point Number System (BFP), Dynamic Fixed-Point Number System (DFXP) and Posit Number System (PNS). The authors explore the impact of these number systems on the performance and hardware design of DNNs, highlighting the challenges associated with each number system and various solutions that are proposed for addressing them Bibliography Includes bibliographical references Notes Online resource; title from PDF title page (SpringerLink, viewed September 8, 2023) Subject Neural networks (Computer science) Neural networks (Computer science) Form Electronic book Author Sakellariou, Vasilis, author Saleh, Hani, author Al-Qutayri, Mahmoud, author Mohammad, Baker, author Stouraitis, Th. (Thanos), author. ISBN 9783031381331 3031381335

  Permalink