{
  "id": "1806.02920",
  "version": "v1",
  "published": "2018-06-07T22:57:16.000Z",
  "updated": "2018-06-07T22:57:16.000Z",
  "title": "GAIN: Missing Data Imputation using Generative Adversarial Nets",
  "authors": [
    "Jinsung Yoon",
    "James Jordon",
    "Mihaela van der Schaar"
  ],
  "comment": "10 pages, 3 figures, 2018 International Conference of Machine Learning",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "We propose a novel method for imputing missing data by adapting the well-known Generative Adversarial Nets (GAN) framework. Accordingly, we call our method Generative Adversarial Imputation Nets (GAIN). The generator (G) observes some components of a real data vector, imputes the missing components conditioned on what is actually observed, and outputs a completed vector. The discriminator (D) then takes a completed vector and attempts to determine which components were actually observed and which were imputed. To ensure that D forces G to learn the desired distribution, we provide D with some additional information in the form of a hint vector. The hint reveals to D partial information about the missingness of the original sample, which is used by D to focus its attention on the imputation quality of particular components. This hint ensures that G does in fact learn to generate according to the true data distribution. We tested our method on various datasets and found that GAIN significantly outperforms state-of-the-art imputation methods.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2018-06-07T22:57:16.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "generative adversarial nets",
      "missing data imputation",
      "outperforms state-of-the-art imputation methods",
      "significantly outperforms state-of-the-art imputation",
      "method generative adversarial imputation nets"
    ],
    "tags": [
      "conference paper"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 10,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}