Deep Learning on Mobile Devices With Neural Processing Units

Tianxiang Tan; Guohong Cao

doi:10.1109/MC.2022.3215780

Deep Learning on Mobile Devices With Neural Processing Units

Tianxiang Tan, Guohong Cao

Computer Science and Engineering

Research output: Contribution to specialist publication › Article

Abstract

Although deep neural networks (DNNs) can require less processing time when running on neural processing units (NPUs) instead of CPUs, their accuracy can degrade. We point out three research directions to improve DNN performance on mobile devices with NPUs.

Original language	English (US)
Pages	48-57
Number of pages	10
Volume	56
No	8
Specialist publication	Computer
DOIs	https://doi.org/10.1109/MC.2022.3215780
State	Published - Aug 1 2023

All Science Journal Classification (ASJC) codes

Computer Science(all)

Access to Document

10.1109/MC.2022.3215780

Cite this

@misc{f7aa05a41f0a4f9f837f978486b3ae16,

title = "Deep Learning on Mobile Devices With Neural Processing Units",

abstract = "Although deep neural networks (DNNs) can require less processing time when running on neural processing units (NPUs) instead of CPUs, their accuracy can degrade. We point out three research directions to improve DNN performance on mobile devices with NPUs.",

author = "Tianxiang Tan and Guohong Cao",

note = "Publisher Copyright: {\textcopyright} 1970-2012 IEEE.",

year = "2023",

month = aug,

day = "1",

doi = "10.1109/MC.2022.3215780",

language = "English (US)",

volume = "56",

pages = "48--57",

journal = "Computer",

issn = "0018-9162",

publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Deep Learning on Mobile Devices With Neural Processing Units

AU - Tan, Tianxiang

AU - Cao, Guohong

PY - 2023/8/1

Y1 - 2023/8/1

N2 - Although deep neural networks (DNNs) can require less processing time when running on neural processing units (NPUs) instead of CPUs, their accuracy can degrade. We point out three research directions to improve DNN performance on mobile devices with NPUs.

AB - Although deep neural networks (DNNs) can require less processing time when running on neural processing units (NPUs) instead of CPUs, their accuracy can degrade. We point out three research directions to improve DNN performance on mobile devices with NPUs.

UR - http://www.scopus.com/inward/record.url?scp=85167662566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85167662566&partnerID=8YFLogxK

U2 - 10.1109/MC.2022.3215780

DO - 10.1109/MC.2022.3215780

M3 - Article

AN - SCOPUS:85167662566

SN - 0018-9162

VL - 56

SP - 48

EP - 57

JO - Computer

JF - Computer

ER -

Deep Learning on Mobile Devices With Neural Processing Units

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Cite this