{
  "id": "1912.03036",
  "version": "v1",
  "published": "2019-12-06T09:24:56.000Z",
  "updated": "2019-12-06T09:24:56.000Z",
  "title": "Improved PAC-Bayesian Bounds for Linear Regression",
  "authors": [
    "Vera Shalaeva",
    "Alireza Fakhrizadeh Esfahani",
    "Pascal Germain",
    "Mihaly Petreczky"
  ],
  "journal": "Thirty-Fourth AAAI Conference on Artificial Intelligence, Feb 2020, New York, United States",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "In this paper, we improve the PAC-Bayesian error bound for linear regression derived in Germain et al. [10]. The improvements are twofold. First, the proposed error bound is tighter, and converges to the generalization loss with a well-chosen temperature parameter. Second, the error bound also holds for training data that are not independently sampled. In particular, the error bound applies to certain time series generated by well-known classes of dynamical models, such as ARX models.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2019-12-06T09:24:56.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "linear regression",
      "pac-bayesian bounds",
      "pac-bayesian error bound",
      "error bound applies",
      "well-chosen temperature parameter"
    ],
    "tags": [
      "journal article"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}