{
  "id": "1903.07378",
  "version": "v1",
  "published": "2019-03-18T12:09:36.000Z",
  "updated": "2019-03-18T12:09:36.000Z",
  "title": "On-line learning dynamics of ReLU neural networks using statistical physics techniques",
  "authors": [
    "Michiel Straat",
    "Michael Biehl"
  ],
  "comment": "Accepted contribution: ESANN 2019, 6 pages European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning 2019",
  "categories": [
    "cs.LG",
    "cond-mat.dis-nn",
    "stat.ML"
  ],
  "abstract": "We introduce exact macroscopic on-line learning dynamics of two-layer neural networks with ReLU units in the form of a system of differential equations, using techniques borrowed from statistical physics. For the first experiments, numerical solutions reveal similar behavior compared to sigmoidal activation researched in earlier work. In these experiments the theoretical results show good correspondence with simulations. In ove-rrealizable and unrealizable learning scenarios, the learning behavior of ReLU networks shows distinctive characteristics compared to sigmoidal networks.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2019-03-18T12:09:36.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "relu neural networks",
      "statistical physics techniques",
      "solutions reveal similar behavior",
      "macroscopic on-line learning dynamics"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 6,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}