Advanced search
2 files | 7.77 MB

The cascading neural network : building the Internet of Smart Things

Sam Leroux (UGent) , Steven Bohez (UGent) , Elias De Coninck (UGent) , Tim Verbelen (UGent) , Bert Vankeirsbilck (UGent) , Pieter Simoens (UGent) and Bart Dhoedt (UGent)
Author
Organization
Abstract
Most of the research on deep neural networks so far has been focused on obtaining higher accuracy levels by building increasingly large and deep architectures. Training and evaluating these models is only feasible when large amounts of resources such as processing power and memory are available. Typical applications that could benefit from these models are, however, executed on resource-constrained devices. Mobile devices such as smartphones already use deep learning techniques, but they often have to perform all processing on a remote cloud. We propose a new architecture called a cascading network that is capable of distributing a deep neural network between a local device and the cloud while keeping the required communication network traffic to a minimum. The network begins processing on the constrained device, and only relies on the remote part when the local part does not provide an accurate enough result. The cascading network allows for an early-stopping mechanism during the recall phase of the network. We evaluated our approach in an Internet of Things context where a deep neural network adds intelligence to a large amount of heterogeneous connected devices. This technique enables a whole variety of autonomous systems where sensors, actuators and computing nodes can work together. We show that the cascading architecture allows for a substantial improvement in evaluation speed on constrained devices while the loss in accuracy is kept to a minimum.
Keywords
IBCN, Neural networks, Internet of Things (IoT), Deep learning, Distributed systems and applications, Cloud computing, Mobile systems, Ubiquitous and pervasive computing

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.71 MB
  • 6963 i.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 6.06 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Leroux, Sam, Steven Bohez, Elias De Coninck, Tim Verbelen, Bert Vankeirsbilck, Pieter Simoens, and Bart Dhoedt. 2017. “The Cascading Neural Network : Building the Internet of Smart Things.” Knowledge and Information Systems 52 (3): 791–814.
APA
Leroux, S., Bohez, S., De Coninck, E., Verbelen, T., Vankeirsbilck, B., Simoens, P., & Dhoedt, B. (2017). The cascading neural network : building the Internet of Smart Things. KNOWLEDGE AND INFORMATION SYSTEMS, 52(3), 791–814.
Vancouver
1.
Leroux S, Bohez S, De Coninck E, Verbelen T, Vankeirsbilck B, Simoens P, et al. The cascading neural network : building the Internet of Smart Things. KNOWLEDGE AND INFORMATION SYSTEMS. 2017;52(3):791–814.
MLA
Leroux, Sam, Steven Bohez, Elias De Coninck, et al. “The Cascading Neural Network : Building the Internet of Smart Things.” KNOWLEDGE AND INFORMATION SYSTEMS 52.3 (2017): 791–814. Print.
@article{8533435,
  abstract     = {Most of the research on deep neural networks so far has been focused on obtaining higher accuracy levels by building increasingly large and deep architectures. Training and evaluating these models is only feasible when large amounts of resources such as processing power and memory are available. Typical applications that could benefit from these models are, however, executed on resource-constrained devices. Mobile devices such as smartphones already use deep learning techniques, but they often have to perform all processing on a remote cloud. We propose a new architecture called a cascading network that is capable of distributing a deep neural network between a local device and the cloud while keeping the required communication network traffic to a minimum. The network begins processing on the constrained device, and only relies on the remote part when the local part does not provide an accurate enough result. The cascading network allows for an early-stopping mechanism during the recall phase of the network. We evaluated our approach in an Internet of Things context where a deep neural network adds intelligence to a large amount of heterogeneous connected devices. This technique enables a whole variety of autonomous systems where sensors, actuators and computing nodes can work together. We show that the cascading architecture allows for a substantial improvement in evaluation speed on constrained devices while the loss in accuracy is kept to a minimum.},
  author       = {Leroux, Sam and Bohez, Steven and De Coninck, Elias and Verbelen, Tim and Vankeirsbilck, Bert and Simoens, Pieter and Dhoedt, Bart},
  issn         = {0219-1377},
  journal      = {KNOWLEDGE AND INFORMATION SYSTEMS},
  keyword      = {IBCN,Neural networks,Internet of Things (IoT),Deep learning,Distributed systems and applications,Cloud computing,Mobile systems,Ubiquitous and pervasive computing},
  language     = {eng},
  number       = {3},
  pages        = {791--814},
  title        = {The cascading neural network : building the Internet of Smart Things},
  url          = {http://dx.doi.org/10.1007/s10115-017-1029-1},
  volume       = {52},
  year         = {2017},
}

Altmetric
View in Altmetric
Web of Science
Times cited: