{
  "id": "2211.15880",
  "version": "v1",
  "published": "2022-11-29T02:33:56.000Z",
  "updated": "2022-11-29T02:33:56.000Z",
  "title": "Mirror descent of Hopfield model",
  "authors": [
    "Hyungjoon Soh",
    "Dongyeob Kim",
    "Juno Hwang",
    "Junghyo Jo"
  ],
  "comment": "3 figures",
  "categories": [
    "cs.LG",
    "math.OC"
  ],
  "abstract": "Mirror descent is a gradient descent method that uses a dual space of parametric models. The great idea has been developed in convex optimization, but not yet widely applied in machine learning. In this study, we provide a possible way that the mirror descent can help data-driven parameter initialization of neural networks. We adopt the Hopfield model as a prototype of neural networks, we demonstrate that the mirror descent can train the model more effectively than the usual gradient descent with random parameter initialization.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2022-11-29T02:33:56.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "mirror descent",
      "hopfield model",
      "neural networks",
      "help data-driven parameter initialization",
      "random parameter initialization"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}