{
  "annotation_id": "17aecbb6-0aed-4c68-a2d2-792bd2d1abf5",
  "date_created": "2026-03-02T18:00:49.854000Z",
  "date_modified": "2026-03-02T18:00:49.854000Z",
  "file_hash": "92a25d83cdbf05abce022ca639c956d4c8ed670a163ae673313bd6854047f944",
  "private": false,
  "record": {
    "abstract": "We investigate supervised learning in neural networks. We consider a\nmulti-layered feed-forward network with back propagation. We find that the\nnetwork of small-world connectivity reduces the learning error and learning\ntime when compared to the networks of regular or random connectivity. Our study\nhas potential applications in the domain of data-mining, image processing,\nspeech recognition, and pattern recognition.",
    "arxiv_id": "physics/0402076",
    "authors": [
      "D. Simard",
      "L. Nadeau",
      "H. Kr\u00f6ger"
    ],
    "categories": [
      "physics.bio-ph",
      "cond-mat.dis-nn"
    ],
    "doi": "10.1016/j.physleta.2004.12.078",
    "journal_ref": "Phys. Lett. A336 (2005) 8-15.",
    "title": "Fastest learning in small world neural networks",
    "url": "https://arxiv.org/abs/physics/0402076"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "a7a76c03-bb43-4cbd-ad14-72c1a00ffc38",
    "id": "arXiv Dataset IDs",
    "type": "Model",
    "variant": "snapshot-2026-03-01",
    "version": "0.1.0"
  },
  "user_id": 1000002
}