Scalable Gamma-Driven Multilayer Network for Brain Workload Detection Through Functional Near-Infrared Spectroscopy

Edmond Q. Wu; Zhiri Tang; Yuxuan Yao; Xu Yi Qiu; Ping Yu Deng; Pengwen Xiong; Aiguo Song; Li Min Zhu; Meng Chu Zhou

doi:10.1109/TCYB.2021.3116964

Scalable Gamma-Driven Multilayer Network for Brain Workload Detection Through Functional Near-Infrared Spectroscopy

Edmond Q. Wu, Zhiri Tang, Yuxuan Yao, Xu Yi Qiu, Ping Yu Deng, Pengwen Xiong, Aiguo Song, Li Min Zhu, Meng Chu Zhou

Electrical and Computer Engineering

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

1 Scopus citations

Abstract

This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.

Original language	English (US)
Journal	IEEE Transactions on Cybernetics
DOIs	https://doi.org/10.1109/TCYB.2021.3116964
State	Accepted/In press - 2021

All Science Journal Classification (ASJC) codes

Software
Control and Systems Engineering
Information Systems
Human-Computer Interaction
Computer Science Applications
Electrical and Electronic Engineering

Keywords

Blood
Brain modeling
Brain workload
cognitive state
Fatigue
Gamma distribution
Matrix decomposition
Neurons
non-negative matrix network
Nonhomogeneous media
pilots' fatigue
Wigner-Ville distribution.

Access to Document

10.1109/TCYB.2021.3116964

Cite this

@article{ed224c4a810e4728be4e57d2309154ef,

title = "Scalable Gamma-Driven Multilayer Network for Brain Workload Detection Through Functional Near-Infrared Spectroscopy",

abstract = "This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.",

keywords = "Blood, Brain modeling, Brain workload, cognitive state, Fatigue, Gamma distribution, Matrix decomposition, Neurons, non-negative matrix network, Nonhomogeneous media, pilots' fatigue, Wigner-Ville distribution.",

author = "Wu, {Edmond Q.} and Zhiri Tang and Yuxuan Yao and Qiu, {Xu Yi} and Deng, {Ping Yu} and Pengwen Xiong and Aiguo Song and Zhu, {Li Min} and Zhou, {Meng Chu}",

note = "Publisher Copyright: IEEE",

year = "2021",

doi = "10.1109/TCYB.2021.3116964",

language = "English (US)",

journal = "IEEE Transactions on Cybernetics",

issn = "2168-2267",

publisher = "IEEE Advancing Technology for Humanity",

}

TY - JOUR

T1 - Scalable Gamma-Driven Multilayer Network for Brain Workload Detection Through Functional Near-Infrared Spectroscopy

AU - Wu, Edmond Q.

AU - Tang, Zhiri

AU - Yao, Yuxuan

AU - Qiu, Xu Yi

AU - Deng, Ping Yu

AU - Xiong, Pengwen

AU - Song, Aiguo

AU - Zhu, Li Min

AU - Zhou, Meng Chu

N1 - Publisher Copyright: IEEE

PY - 2021

Y1 - 2021

N2 - This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.

AB - This work proposes a scalable gamma non-negative matrix network (SGNMN), which uses a Poisson randomized Gamma factor analysis to obtain the neurons of the first layer of a network. These neurons obey Gamma distribution whose shape parameter infers the neurons of the next layer of the network and their related weights. Upsampling the connection weights follows a Dirichlet distribution. Downsampling hidden units obey Gamma distribution. This work performs up-down sampling on each layer to learn the parameters of SGNMN. Experimental results indicate that the width and depth of SGNMN are closely related, and a reasonable network structure for accurately detecting brain fatigue through functional near-infrared spectroscopy can be obtained by considering network width, depth, and parameters.

KW - Blood

KW - Brain modeling

KW - Brain workload

KW - cognitive state

KW - Fatigue

KW - Gamma distribution

KW - Matrix decomposition

KW - Neurons

KW - non-negative matrix network

KW - Nonhomogeneous media

KW - pilots' fatigue

KW - Wigner-Ville distribution.

UR - http://www.scopus.com/inward/record.url?scp=85118585201&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85118585201&partnerID=8YFLogxK

U2 - 10.1109/TCYB.2021.3116964

DO - 10.1109/TCYB.2021.3116964

M3 - Article

AN - SCOPUS:85118585201

JO - IEEE Transactions on Cybernetics

JF - IEEE Transactions on Cybernetics

SN - 2168-2267

ER -

Scalable Gamma-Driven Multilayer Network for Brain Workload Detection Through Functional Near-Infrared Spectroscopy

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this