{
  "id": "2010.12711",
  "version": "v1",
  "published": "2020-10-23T23:41:03.000Z",
  "updated": "2020-10-23T23:41:03.000Z",
  "title": "On Convergence and Generalization of Dropout Training",
  "authors": [
    "Poorya Mianjy",
    "Raman Arora"
  ],
  "journal": "In Proceedings of Advances in Neural Information Processing Systems (NeurIPS), 2020",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "We study dropout in two-layer neural networks with rectified linear unit (ReLU) activations. Under mild overparametrization and assuming that the limiting kernel can separate the data distribution with a positive margin, we show that dropout training with logistic loss achieves $\\epsilon$-suboptimality in test error in $O(1/\\epsilon)$ iterations.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-10-23T23:41:03.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "dropout training",
      "convergence",
      "generalization",
      "two-layer neural networks",
      "logistic loss achieves"
    ],
    "tags": [
      "journal article"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}