{
  "id": "2006.10236",
  "version": "v1",
  "published": "2020-06-18T02:10:56.000Z",
  "updated": "2020-06-18T02:10:56.000Z",
  "title": "Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Models",
  "authors": [
    "Siavash Khodadadeh",
    "Sharare Zehtabian",
    "Saeed Vahidian",
    "Weijia Wang",
    "Bill Lin",
    "Ladislau Bölöni"
  ],
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "Unsupervised meta-learning approaches rely on synthetic meta-tasks that are created using techniques such as random selection, clustering and/or augmentation. Unfortunately, clustering and augmentation are domain-dependent, and thus they require either manual tweaking or expensive learning. In this work, we describe an approach that generates meta-tasks using generative models. A critical component is a novel approach of sampling from the latent space that generates objects grouped into synthetic classes forming the training and validation data of a meta-task. We find that the proposed approach, LAtent Space Interpolation Unsupervised Meta-learning (LASIUM), outperforms or is competitive with current unsupervised learning baselines on few-shot classification tasks on the most widely used benchmark datasets. In addition, the approach promises to be applicable without manual tweaking over a wider range of domains than previous approaches.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-06-18T02:10:56.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "generative models",
      "latent-space interpolation",
      "space interpolation unsupervised meta-learning",
      "latent space interpolation",
      "few-shot classification tasks"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}