Conference paper Open Access

Efficient Winograd-based Convolution Kernel Implementation on Edge Devices

Xygkis, Athanasios; Papadopoulos, Lazaros; Moloney, David; Soudris, Dimitrios; Yous, Sofiane


JSON-LD (schema.org) Export

{
  "description": "<p>The implementation of Convolutional Neural Networks on edge Internet of Things (IoT) devices is a significant programming challenge, due to the limited computational resources and the real-time requirements of modern applications. This work focuses on the efficient implementation of the Winograd convolution, based on a set of application-independent and Winograd-specific software techniques for improving the utilization of the edge devices computational resources. The proposed techniques were evaluated in Intel/Movidius Myriad2 platform, using 4 CNNs of various computational requirements. The results show significant performance improvements, up to 54%, over other convolution algorithms.</p>", 
  "license": "https://creativecommons.org/licenses/by/4.0/legalcode", 
  "creator": [
    {
      "affiliation": "Intel Corporation, Ireland", 
      "@type": "Person", 
      "name": "Xygkis, Athanasios"
    }, 
    {
      "affiliation": "School of ECE, NTUA, Greece", 
      "@type": "Person", 
      "name": "Papadopoulos, Lazaros"
    }, 
    {
      "affiliation": "Intel Corporation, Ireland", 
      "@type": "Person", 
      "name": "Moloney, David"
    }, 
    {
      "affiliation": "School of ECE, NTUA, Greece", 
      "@type": "Person", 
      "name": "Soudris, Dimitrios"
    }, 
    {
      "affiliation": "Intel Corporation, Ireland", 
      "@type": "Person", 
      "name": "Yous, Sofiane"
    }
  ], 
  "headline": "Efficient Winograd-based Convolution Kernel Implementation on Edge Devices", 
  "image": "https://zenodo.org/static/img/logos/zenodo-gradient-round.svg", 
  "datePublished": "2018-09-20", 
  "url": "https://zenodo.org/record/3347180", 
  "@context": "https://schema.org/", 
  "identifier": "https://doi.org/10.1145/3195970.3196041", 
  "@id": "https://doi.org/10.1145/3195970.3196041", 
  "@type": "ScholarlyArticle", 
  "name": "Efficient Winograd-based Convolution Kernel Implementation on Edge Devices"
}
35
1,043
views
downloads
Views 35
Downloads 1,043
Data volume 365.7 MB
Unique views 33
Unique downloads 1,012

Share

Cite as