Internet of Things. IoT Infrastructures. Second International Summit, IoT 360° 2015, Rome, Italy, October 27-29, 2015, Revised Selected Papers, Part II

Research Article

Distributed Neural Networks for Internet of Things: The Big-Little Approach

Download
262 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-47075-7_52,
        author={Elias Coninck and Tim Verbelen and Bert Vankeirsbilck and Steven Bohez and Pieter Simoens and Piet Demeester and Bart Dhoedt},
        title={Distributed Neural Networks for Internet of Things: The Big-Little Approach},
        proceedings={Internet of Things. IoT Infrastructures. Second International Summit, IoT 360° 2015, Rome, Italy, October 27-29, 2015, Revised Selected Papers, Part II},
        proceedings_a={IOT360},
        year={2017},
        month={6},
        keywords={Deep neural networks Distributed intelligence Internet of things},
        doi={10.1007/978-3-319-47075-7_52}
    }
    
  • Elias Coninck
    Tim Verbelen
    Bert Vankeirsbilck
    Steven Bohez
    Pieter Simoens
    Piet Demeester
    Bart Dhoedt
    Year: 2017
    Distributed Neural Networks for Internet of Things: The Big-Little Approach
    IOT360
    Springer
    DOI: 10.1007/978-3-319-47075-7_52
Elias Coninck1,*, Tim Verbelen1, Bert Vankeirsbilck1, Steven Bohez1, Pieter Simoens1, Piet Demeester1, Bart Dhoedt1
  • 1: Ghent University – iMinds
*Contact email: elias.deconinck@intec.ugent.be

Abstract

Nowadays deep neural networks are widely used to accurately classify input data. An interesting application area is the Internet of Things (IoT), where a massive amount of sensor data has to be classified. The processing power of the cloud is attractive, however the variable latency imposes a major drawback in situations where near real-time classification is required. In order to exploit the apparent trade-off between utilizing the stable but limited embedded computing power of IoT devices and the seemingly unlimited computing power of Cloud computing at the cost of higher and variable latency, we propose a Big-Little architecture for deep neural networks. A small neural network trained to a subset of prioritized output classes is running on the embedded device, while a more specific classification is calculated when required by a large neural network in the cloud. We show the applicability of this concept in the IoT domain by evaluating our approach for state of the art neural network classification problems on popular embedded devices such as the Raspberry Pi and Intel Edison.