{
  "id": "2003.11539",
  "version": "v1",
  "published": "2020-03-25T17:58:42.000Z",
  "updated": "2020-03-25T17:58:42.000Z",
  "title": "Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?",
  "authors": [
    "Yonglong Tian",
    "Yue Wang",
    "Dilip Krishnan",
    "Joshua B. Tenenbaum",
    "Phillip Isola"
  ],
  "comment": "First two authors contributed equally. Code: http://github.com/WangYueFt/rfs/",
  "categories": [
    "cs.CV",
    "cs.LG"
  ],
  "abstract": "The focus of recent meta-learning research has been on the development of learning algorithms that can quickly adapt to test time tasks with limited data and low computational cost. Few-shot learning is widely used as one of the standard benchmarks in meta-learning. In this work, we show that a simple baseline: learning a supervised or self-supervised representation on the meta-training set, followed by training a linear classifier on top of this representation, outperforms state-of-the-art few-shot learning methods. An additional boost can be achieved through the use of self-distillation. This demonstrates that using a good learned embedding model can be more effective than sophisticated meta-learning algorithms. We believe that our findings motivate a rethinking of few-shot image classification benchmarks and the associated role of meta-learning algorithms. Code is available at: http://github.com/WangYueFt/rfs/.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-03-25T17:58:42.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "rethinking few-shot image classification",
      "outperforms state-of-the-art few-shot learning methods",
      "few-shot image classification benchmarks",
      "test time tasks",
      "low computational cost"
    ],
    "tags": [
      "github project"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}