{
  "id": "2207.01996",
  "version": "v1",
  "published": "2022-07-05T12:28:13.000Z",
  "updated": "2022-07-05T12:28:13.000Z",
  "title": "Correlation between entropy and generalizability in a neural network",
  "authors": [
    "Ge Zhang"
  ],
  "categories": [
    "cond-mat.stat-mech",
    "cs.LG"
  ],
  "abstract": "Although neural networks can solve very complex machine-learning problems, the theoretical reason for their generalizability is still not fully understood. Here we use Wang-Landau Mote Carlo algorithm to calculate the entropy (logarithm of the volume of a part of the parameter space) at a given test accuracy, and a given training loss function value or training accuracy. Our results show that entropical forces help generalizability. Although our study is on a very simple application of neural networks (a spiral dataset and a small, fully-connected neural network), our approach should be useful in explaining the generalizability of more complicated neural networks in future works.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2022-07-05T12:28:13.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "correlation",
      "wang-landau mote carlo algorithm",
      "training loss function value",
      "entropical forces help generalizability",
      "test accuracy"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}