STAR: A Mixed Analog Stochastic In-DRAM Convolutional Neural Network Accelerator

Salma Afifi; Ishan Thakkar; Sudeep Pasricha

doi:10.1109/MDAT.2024.3447580

STAR: A Mixed Analog Stochastic In-DRAM Convolutional Neural Network Accelerator

Salma Afifi
, Ishan Thakkar
, Sudeep Pasricha

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Editor's notes: This article addresses the optimization of data movement in accelerating machine learning workloads, one of the most critical issues of state-ofthe- art computing platforms. It presents a novel in-DRAM accelerator for convolutional neural networks using mixed analog-stochastic optimizations and shows significant energy-efficiency improvements. - Umit Ogras, University of Wisconsin, USA.

Original language	English
Pages (from-to)	47-55
Number of pages	9
Journal	IEEE Design and Test
Volume	42
Issue number	1
DOIs	https://doi.org/10.1109/MDAT.2024.3447580
State	Published - 2025

Bibliographical note

Publisher Copyright:
© 2013 IEEE.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

convolution neural networks
in-DRAM processing
stochastic computing

ASJC Scopus subject areas

Software
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/MDAT.2024.3447580

Cite this

@article{fc4c605caa344ccabb8989a2008c994e,

title = "STAR: A Mixed Analog Stochastic In-DRAM Convolutional Neural Network Accelerator",

abstract = "Editor's notes: This article addresses the optimization of data movement in accelerating machine learning workloads, one of the most critical issues of state-ofthe- art computing platforms. It presents a novel in-DRAM accelerator for convolutional neural networks using mixed analog-stochastic optimizations and shows significant energy-efficiency improvements. - Umit Ogras, University of Wisconsin, USA.",

keywords = "convolution neural networks, in-DRAM processing, stochastic computing",

author = "Salma Afifi and Ishan Thakkar and Sudeep Pasricha",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2025",

doi = "10.1109/MDAT.2024.3447580",

language = "English",

volume = "42",

pages = "47--55",

number = "1",

}

TY - JOUR

T1 - STAR

T2 - A Mixed Analog Stochastic In-DRAM Convolutional Neural Network Accelerator

AU - Afifi, Salma

AU - Thakkar, Ishan

AU - Pasricha, Sudeep

PY - 2025

Y1 - 2025

N2 - Editor's notes: This article addresses the optimization of data movement in accelerating machine learning workloads, one of the most critical issues of state-ofthe- art computing platforms. It presents a novel in-DRAM accelerator for convolutional neural networks using mixed analog-stochastic optimizations and shows significant energy-efficiency improvements. - Umit Ogras, University of Wisconsin, USA.

AB - Editor's notes: This article addresses the optimization of data movement in accelerating machine learning workloads, one of the most critical issues of state-ofthe- art computing platforms. It presents a novel in-DRAM accelerator for convolutional neural networks using mixed analog-stochastic optimizations and shows significant energy-efficiency improvements. - Umit Ogras, University of Wisconsin, USA.

KW - convolution neural networks

KW - in-DRAM processing

KW - stochastic computing

UR - https://www.scopus.com/pages/publications/85202722158

UR - https://www.scopus.com/pages/publications/85202722158#tab=citedBy

U2 - 10.1109/MDAT.2024.3447580

DO - 10.1109/MDAT.2024.3447580

M3 - Article

AN - SCOPUS:85202722158

SN - 2168-2356

VL - 42

SP - 47

EP - 55

JO - IEEE Design and Test

JF - IEEE Design and Test

IS - 1

ER -

STAR: A Mixed Analog Stochastic In-DRAM Convolutional Neural Network Accelerator

Abstract

Bibliographical note

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this