{
  "id": "2102.04593",
  "version": "v1",
  "published": "2021-02-09T01:13:36.000Z",
  "updated": "2021-02-09T01:13:36.000Z",
  "title": "Regularized Generative Adversarial Network",
  "authors": [
    "Gabriele Di Cerbo",
    "Ali Hirsa",
    "Ahmad Shayaan"
  ],
  "comment": "18 pages. Comments are welcome!",
  "categories": [
    "cs.LG",
    "cs.AI",
    "cs.CV"
  ],
  "abstract": "We propose a framework for generating samples from a probability distribution that differs from the probability distribution of the training set. We use an adversarial process that simultaneously trains three networks, a generator and two discriminators. We refer to this new model as regularized generative adversarial network (RegGAN). We evaluate RegGAN on a synthetic dataset composed of gray scale images and we further show that it can be used to learn some pre-specified notions in topology (basic topology properties). The work is motivated by practical problems encountered while using generative methods in the art world.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2021-02-09T01:13:36.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "regularized generative adversarial network",
      "probability distribution",
      "gray scale images",
      "basic topology properties",
      "art world"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 18,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}