PASS: Exploiting post-activation sparsity in streaming architectures for CNN acceleration

Alexander Montgomerie-Corcoran, Zhewen Yu, Jianyi Cheng, Christos Savvas Bouganis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

With the ever-growing popularity of Artificial Intelligence, there is an increasing demand for more performant and efficient underlying hardware. Convolutional Neural Networks (CNN) are a workload of particular importance, which achieve high accuracy in computer vision applications. Inside CNNs, a significant number of the post-activation values are zero, resulting in many redundant computations. Recent works have explored this post-activation sparsity on instruction-based CNN accelerators but not on streaming CNN accelerators, despite the fact that streaming architectures are considered the leading design methodology in terms of performance. In this paper, we highlight the challenges associated with exploiting post-activation sparsity for performance gains in streaming CNN accelerators, and demonstrate our approach to address them. Using a set of modern CNN benchmarks, our streaming sparse accelerators achieve 1.41 x to 1.93 x efficiency (GOP/sDSP) compared to state-of-the-art instruction-based sparse accelerators.
Original languageEnglish
Title of host publication2023 33rd International Conference on Field-Programmable Logic and Applications
EditorsIoannis Sourdis, Nele Mentens, Leonel Sousa, Pedro Trancoso
PublisherInstitute of Electrical and Electronics Engineers
Pages288-293
Number of pages6
ISBN (Electronic)9798350341515
ISBN (Print)9798350341522
DOIs
Publication statusPublished - 2 Nov 2023
Event33rd International Conference on Field-Programmable Logic and Applications - Gothenburg, Sweden
Duration: 4 Sept 20238 Sept 2023

Publication series

NameProceedings of the International Conference on Field-Programmable Logic and Applications
PublisherInstitute of Electrical and Electronics Engineers
ISSN (Print)1946-147X
ISSN (Electronic)1946-1488

Conference

Conference33rd International Conference on Field-Programmable Logic and Applications
Abbreviated titleFLP 2023
Country/TerritorySweden
CityGothenburg
Period4/09/238/09/23

Fingerprint

Dive into the research topics of 'PASS: Exploiting post-activation sparsity in streaming architectures for CNN acceleration'. Together they form a unique fingerprint.

Cite this