Hardware and Software Optimizations for Capsule Networks

Alberto Marchisio; Beatrice Bussolino; Alessio Colucci; Vojtech Mrazek; Muhammad Abdullah Hanif; Maurizio Martina; Guido Masera; Muhammad Shafique

doi:10.1007/978-3-031-39932-12

Hardware and Software Optimizations for Capsule Networks

Alberto Marchisio, Beatrice Bussolino, Alessio Colucci, Vojtech Mrazek, Muhammad Abdullah Hanif, Maurizio Martina, Guido Masera, Muhammad Shafique

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

Among advanced Deep Neural Network models, Capsule Networks (CapsNets) have shown high learning and generalization capabilities for advanced tasks. Their capability to learn hierarchical information of features makes them appealing in many applications. However, their compute-intensive nature poses several challenges for their deployment on resource-constrained devices. This chapter provides an optimization flow at the software and at the hardware level for improving the energy efficiency of the CapsNets’ execution.

Original language	English (US)
Title of host publication	Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing
Subtitle of host publication	Software Optimizations and Hardware/Software Codesign
Publisher	Springer Nature
Pages	303-328
Number of pages	26
ISBN (Electronic)	9783031399329
ISBN (Print)	9783031399312
DOIs	https://doi.org/10.1007/978-3-031-39932-12
State	Published - Jan 1 2023

Keywords

Capsule networks
Deep learning
Energy efficiency
Hardware accelerator
Software optimizations

ASJC Scopus subject areas

General Computer Science
General Engineering
General Social Sciences

Access to Document

10.1007/978-3-031-39932-12

Cite this

Marchisio, A., Bussolino, B., Colucci, A., Mrazek, V., Hanif, M. A., Martina, M., Masera, G., & Shafique, M. (2023). Hardware and Software Optimizations for Capsule Networks. In Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing: Software Optimizations and Hardware/Software Codesign (pp. 303-328). Springer Nature. https://doi.org/10.1007/978-3-031-39932-12

@inbook{d55a13755bd14fa9ad30870e111f307a,

title = "Hardware and Software Optimizations for Capsule Networks",

abstract = "Among advanced Deep Neural Network models, Capsule Networks (CapsNets) have shown high learning and generalization capabilities for advanced tasks. Their capability to learn hierarchical information of features makes them appealing in many applications. However, their compute-intensive nature poses several challenges for their deployment on resource-constrained devices. This chapter provides an optimization flow at the software and at the hardware level for improving the energy efficiency of the CapsNets{\textquoteright} execution.",

keywords = "Capsule networks, Deep learning, Energy efficiency, Hardware accelerator, Software optimizations",

author = "Alberto Marchisio and Beatrice Bussolino and Alessio Colucci and Vojtech Mrazek and Hanif, {Muhammad Abdullah} and Maurizio Martina and Guido Masera and Muhammad Shafique",

note = "Publisher Copyright: {\textcopyright} 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG.",

year = "2023",

month = jan,

day = "1",

doi = "10.1007/978-3-031-39932-12",

language = "English (US)",

isbn = "9783031399312",

pages = "303--328",

booktitle = "Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing",

publisher = "Springer Nature",

address = "United States",

}

TY - CHAP

T1 - Hardware and Software Optimizations for Capsule Networks

AU - Marchisio, Alberto

AU - Bussolino, Beatrice

AU - Colucci, Alessio

AU - Mrazek, Vojtech

AU - Hanif, Muhammad Abdullah

AU - Martina, Maurizio

AU - Masera, Guido

AU - Shafique, Muhammad

PY - 2023/1/1

Y1 - 2023/1/1

N2 - Among advanced Deep Neural Network models, Capsule Networks (CapsNets) have shown high learning and generalization capabilities for advanced tasks. Their capability to learn hierarchical information of features makes them appealing in many applications. However, their compute-intensive nature poses several challenges for their deployment on resource-constrained devices. This chapter provides an optimization flow at the software and at the hardware level for improving the energy efficiency of the CapsNets’ execution.

AB - Among advanced Deep Neural Network models, Capsule Networks (CapsNets) have shown high learning and generalization capabilities for advanced tasks. Their capability to learn hierarchical information of features makes them appealing in many applications. However, their compute-intensive nature poses several challenges for their deployment on resource-constrained devices. This chapter provides an optimization flow at the software and at the hardware level for improving the energy efficiency of the CapsNets’ execution.

KW - Capsule networks

KW - Deep learning

KW - Energy efficiency

KW - Hardware accelerator

KW - Software optimizations

UR - http://www.scopus.com/inward/record.url?scp=85196470066&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85196470066&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-39932-12

DO - 10.1007/978-3-031-39932-12

M3 - Chapter

AN - SCOPUS:85196470066

SN - 9783031399312

SP - 303

EP - 328

BT - Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing

PB - Springer Nature

ER -

Hardware and Software Optimizations for Capsule Networks

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this