Conference paper Open Access

Training Convolutional Neural Networks with Competitive Hebbian Learning Approaches

Gabriele Lagani; Fabrizio Falchi; Claudio Gennaro; Giuseppe Amato


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/dc8f41f3-f963-4b02-9365-8497b31e277b/ACAIN2021_paper_20.pdf"
      }, 
      "checksum": "md5:0e518291c88a6e2eb41f885e3885ba3b", 
      "bucket": "dc8f41f3-f963-4b02-9365-8497b31e277b", 
      "key": "ACAIN2021_paper_20.pdf", 
      "type": "pdf", 
      "size": 519723
    }
  ], 
  "owners": [
    112563
  ], 
  "doi": "10.1007/978-3-030-95467-3_2", 
  "stats": {
    "version_unique_downloads": 30.0, 
    "unique_views": 21.0, 
    "views": 27.0, 
    "version_views": 27.0, 
    "unique_downloads": 30.0, 
    "version_unique_views": 21.0, 
    "volume": 15591690.0, 
    "version_downloads": 30.0, 
    "downloads": 30.0, 
    "version_volume": 15591690.0
  }, 
  "links": {
    "doi": "https://doi.org/10.1007/978-3-030-95467-3_2", 
    "latest_html": "https://zenodo.org/record/6367135", 
    "bucket": "https://zenodo.org/api/files/dc8f41f3-f963-4b02-9365-8497b31e277b", 
    "badge": "https://zenodo.org/badge/doi/10.1007/978-3-030-95467-3_2.svg", 
    "html": "https://zenodo.org/record/6367135", 
    "latest": "https://zenodo.org/api/records/6367135"
  }, 
  "created": "2022-03-18T09:14:37.536178+00:00", 
  "updated": "2022-03-20T01:49:22.247071+00:00", 
  "conceptrecid": "6367134", 
  "revision": 5, 
  "id": 6367135, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.1007/978-3-030-95467-3_2", 
    "description": "<p>We explore competitive Hebbian learning strategies to train feature detectors in Convolutional Neural Networks (CNNs), without supervision. We consider variants of the Winner-Takes-All (WTA) strategy explored in previous works, i.e. k-WTA, e-soft-WTA and p-soft-WTA, performing experiments on different object recognition datasets. Results suggest that the Hebbian approaches are effective to train early feature extraction layers, or to re-train higher layers of a pre-trained network, with soft competition generally performing better than other Hebbian approaches explored in this work. Our findings encourage a path of cooperation between neuroscience and computer science towards a deeper investigation of biologically inspired learning principles.</p>", 
    "language": "eng", 
    "title": "Training Convolutional Neural Networks with Competitive Hebbian Learning Approaches", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "notes": "In Nicosia G. et al. (eds) Machine Learning, Optimization, and Data Science. LOD 2021. Lecture Notes in Computer Science, vol 13163. Springer, Cham. https://doi.org/10.1007/978-3-030-95467-3_2", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "6367134"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "6367135"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "ai4eu"
      }, 
      {
        "id": "ai4media"
      }
    ], 
    "grants": [
      {
        "code": "951911", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::951911"
        }, 
        "title": "A European Excellence Centre for Media, Society and Democracy", 
        "acronym": "AI4Media", 
        "program": "Horizon 2020 Framework Programme - Research and Innovation action", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }, 
      {
        "code": "825619", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::825619"
        }, 
        "title": "A European AI On Demand Platform and Ecosystem", 
        "acronym": "AI4EU", 
        "program": "H2020", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }
    ], 
    "keywords": [
      "Neural networks, Machine learning, Hebbian learning Competitive learning, Computer vision,Biologically inspired"
    ], 
    "publication_date": "2022-02-02", 
    "creators": [
      {
        "affiliation": "University of Pisa", 
        "name": "Gabriele Lagani"
      }, 
      {
        "affiliation": "CNR-ISTI", 
        "name": "Fabrizio Falchi"
      }, 
      {
        "affiliation": "CNR-ISTI", 
        "name": "Claudio Gennaro"
      }, 
      {
        "affiliation": "CNR-ISTI", 
        "name": "Giuseppe Amato"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }
  }
}
27
30
views
downloads
Views 27
Downloads 30
Data volume 15.6 MB
Unique views 21
Unique downloads 30

Share

Cite as