Multiplexer Optimization for Adders in Stochastic Computing

Sercan Aygun, M. Hassan Najafi, Lida Kouhalvandi, Ece Olcay Gunes

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This study presents an optimization algorithm for multiplexer (MUX)-based scaled addition for stochastic computing (SC). Accumulation operation can be performed in SC using a MUX unit. Cascaded structures of 2m-to-1 MUXs are used for the accumulation of multiple terms. Optimizing these designs holds significance in cases of accumulating a large number of inputs. The depth of the cascaded MUXs varies with m, affecting the hardware cost, delay, and accuracy. The proposed algorithm performs stage-wise optimization of m. Evaluation results show a lower hardware cost and a higher accuracy compared to the standard MUX-based SC addition using 2-to-1 MUXs for SC-based neural networks.

Original languageEnglish
Title of host publicationProceedings of the 18th ACM International Symposium on Nanoscale Architectures, NANOARCH 2023
PublisherAssociation for Computing Machinery
ISBN (Electronic)9798400703256
DOIs
Publication statusPublished - 18 Dec 2023
Event18th ACM International Symposium on Nanoscale Architectures, NANOARCH 2023 - Dresden, Germany
Duration: 18 Dec 202320 Dec 2023

Publication series

NameACM International Conference Proceeding Series

Conference

Conference18th ACM International Symposium on Nanoscale Architectures, NANOARCH 2023
Country/TerritoryGermany
CityDresden
Period18/12/2320/12/23

Bibliographical note

Publisher Copyright:
© 2023 ACM.

Funding

This work was supported in part by National Science Foundation (NSF) grant #2019511, the Louisiana Board of Regents Support Fund #LEQSF(2020-23)-RD-A-26, and generous gifts from Cisco, Xilinx, and Nvidia.

FundersFunder number
National Science Foundation2019511
Louisiana Board of Regents2020-23)-RD-A-26

    Keywords

    • Multiplexer
    • neural networks
    • optimization
    • stochastic computing

    Fingerprint

    Dive into the research topics of 'Multiplexer Optimization for Adders in Stochastic Computing'. Together they form a unique fingerprint.

    Cite this