USEFUSE: Uniform stride for enhanced performance in fused layer architecture of deep neural networks

URN to cite this document:: urn:nbn:de:bvb:355-epub-767708
DOI to cite this document:: 10.5283/epub.76770

Ibrahim, Muhammad Sohail ; Usman, Muhammad

; Lee, Jeong-A.

License: Creative Commons Attribution 4.0
PDF - Published Version
(1MB)

Date of publication of this fulltext: 04 Jun 2025 09:15

Alternative links to fulltext:DOI Publisher

Details

Preview

Export bibliographical data

Abstract

Convolutional Neural Networks (CNNs) are crucial in various applications, but deploying them on resource-constrained edge devices poses challenges. This study presents the Sum-of-Products (SOP) units for convolution, which utilize low-latency left-to-right bit-serial arithmetic to minimize response time and enhance overall performance. The study proposes a methodology for fusing multiple ...