Conference paper Open Access

Efficient Winograd-based Convolution Kernel Implementation on Edge Devices

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


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/ee5e138a-e4a1-4c1c-a3c5-c30450f53c59/Efficient_Winograd-based2.pdf"
      }, 
      "checksum": "md5:b23e0fe233b68b39515e3278fdff0b96", 
      "bucket": "ee5e138a-e4a1-4c1c-a3c5-c30450f53c59", 
      "key": "Efficient_Winograd-based2.pdf", 
      "type": "pdf", 
      "size": 350657
    }
  ], 
  "owners": [
    43983
  ], 
  "doi": "10.1145/3195970.3196041", 
  "stats": {
    "version_unique_downloads": 1005.0, 
    "unique_views": 33.0, 
    "views": 35.0, 
    "version_views": 35.0, 
    "unique_downloads": 1005.0, 
    "version_unique_views": 33.0, 
    "volume": 362929995.0, 
    "version_downloads": 1035.0, 
    "downloads": 1035.0, 
    "version_volume": 362929995.0
  }, 
  "links": {
    "doi": "https://doi.org/10.1145/3195970.3196041", 
    "latest_html": "https://zenodo.org/record/3347180", 
    "bucket": "https://zenodo.org/api/files/ee5e138a-e4a1-4c1c-a3c5-c30450f53c59", 
    "badge": "https://zenodo.org/badge/doi/10.1145/3195970.3196041.svg", 
    "html": "https://zenodo.org/record/3347180", 
    "latest": "https://zenodo.org/api/records/3347180"
  }, 
  "created": "2019-07-23T17:12:04.838678+00:00", 
  "updated": "2020-01-20T17:17:28.768684+00:00", 
  "conceptrecid": "3347179", 
  "revision": 4, 
  "id": 3347180, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.1145/3195970.3196041", 
    "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": {
      "id": "CC-BY-4.0"
    }, 
    "title": "Efficient Winograd-based Convolution Kernel Implementation on Edge Devices", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3347179"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "3347180"
          }
        }
      ]
    }, 
    "grants": [
      {
        "code": "780572", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::780572"
        }, 
        "title": "Software Development toolKit for Energy optimization and technical Debt elimination", 
        "acronym": "SDK4ED", 
        "program": "H2020", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }
    ], 
    "publication_date": "2018-09-20", 
    "creators": [
      {
        "affiliation": "Intel Corporation, Ireland", 
        "name": "Xygkis, Athanasios"
      }, 
      {
        "affiliation": "School of ECE, NTUA, Greece", 
        "name": "Papadopoulos, Lazaros"
      }, 
      {
        "affiliation": "Intel Corporation, Ireland", 
        "name": "Moloney, David"
      }, 
      {
        "affiliation": "School of ECE, NTUA, Greece", 
        "name": "Soudris, Dimitrios"
      }, 
      {
        "affiliation": "Intel Corporation, Ireland", 
        "name": "Yous, Sofiane"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }
  }
}
35
1,035
views
downloads
Views 35
Downloads 1,035
Data volume 362.9 MB
Unique views 33
Unique downloads 1,005

Share

Cite as