{
  "id": "1906.10343",
  "version": "v1",
  "published": "2019-06-25T06:42:05.000Z",
  "updated": "2019-06-25T06:42:05.000Z",
  "title": "Semi-Supervised Learning with Self-Supervised Networks",
  "authors": [
    "Phi Vu Tran"
  ],
  "comment": "initial tech report, 10 pages",
  "categories": [
    "cs.LG",
    "cs.CV",
    "stat.ML"
  ],
  "abstract": "Recent advances in semi-supervised learning have shown tremendous potential in overcoming a major barrier to the success of modern machine learning algorithms: access to vast amounts of human-labeled training data. Algorithms based on self-ensemble learning and virtual adversarial training can harness the abundance of unlabeled data to produce impressive state-of-the-art results on a number of semi-supervised benchmarks, approaching the performance of strong supervised baselines using only a fraction of the available labeled data. However, these methods often require careful tuning of many hyper-parameters and are usually not easy to implement in practice. In this work, we present a conceptually simple yet effective semi-supervised algorithm based on self-supervised learning to combine semantic feature representations from unlabeled data. Our models are efficiently trained end-to-end for the joint, multi-task learning of labeled and unlabeled data in a single stage. Striving for simplicity and practicality, our approach requires no additional hyper-parameters to tune for optimal performance beyond the standard set for training convolutional neural networks. We conduct a comprehensive empirical evaluation of our models for semi-supervised image classification on SVHN, CIFAR-10 and CIFAR-100, and demonstrate results competitive with, and in some cases exceeding, prior state of the art. Reference code and data are available at https://github.com/vuptran/sesemi",
  "revisions": [
    {
      "version": "v1",
      "updated": "2019-06-25T06:42:05.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "semi-supervised learning",
      "self-supervised networks",
      "unlabeled data",
      "training convolutional neural networks",
      "semantic feature representations"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 10,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}