{
  "id": "1911.06515",
  "version": "v1",
  "published": "2019-11-15T08:40:45.000Z",
  "updated": "2019-11-15T08:40:45.000Z",
  "title": "Likelihood Assignment for Out-of-Distribution Inputs in Deep Generative Models is Sensitive to Prior Distribution Choice",
  "authors": [
    "Ryo Kamoi",
    "Kei Kobayashi"
  ],
  "categories": [
    "stat.ML",
    "cs.LG"
  ],
  "abstract": "Recent work has shown that deep generative models assign higher likelihood to out-of-distribution inputs than to training data. We show that a factor underlying this phenomenon is a mismatch between the nature of the prior distribution and that of the data distribution, a problem found in widely used deep generative models such as VAEs and Glow. While a typical choice for a prior distribution is a standard Gaussian distribution, properties of distributions of real data sets may not be consistent with a unimodal prior distribution. This paper focuses on the relationship between the choice of a prior distribution and the likelihoods assigned to out-of-distribution inputs. We propose the use of a mixture distribution as a prior to make likelihoods assigned by deep generative models sensitive to out-of-distribution inputs. Furthermore, we explain the theoretical advantages of adopting a mixture distribution as the prior, and we present experimental results to support our claims. Finally, we demonstrate that a mixture prior lowers the out-of-distribution likelihood with respect to two pairs of real image data sets: Fashion-MNIST vs. MNIST and CIFAR10 vs. SVHN.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2019-11-15T08:40:45.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "deep generative models",
      "out-of-distribution inputs",
      "prior distribution choice",
      "likelihood assignment",
      "generative models assign higher likelihood"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}