{
  "fetched_at": "2026-10-19T10:17:05Z",
  "source_url": "https://www.kaggle.com/competitions/ensta-ssl-cifar10/data",
  "pages": [
    {
      "name": "data-description",
      "content": "\n<p>Please download the dataset CIFAR10 and randomly select 1000 images from the training set. We consider that you have the annotation on these images. You can also select 5000 images equally balance from the training set. This set will be the validation set.</p>\n<p>Finally, the rest of the official training set are images without labels that you can use to do whatever you want.</p>\n<h3>Hence you have four sets.</h3>\n<ul>\n<li>  one training set with the labels (1000 images)</li>\n<li> one validation set with the labels (5000 images) </li>\n<li>  one testing set with the labels (10000 images) (the official testing set)</li>\n<li>one set compose of all the images that you did not select from the official images from CIFAR10. This set is the unlabel set.</li>\n</ul>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p> We want you to be able to train a model when you do not have annotations on all the datasets.</p>\n\n<p>A powerful approach for training such a model on a large amount of data without requiring a large amount of labels is to use semi-supervised learning(SSL) technique. SSL mitigates the requirement for labeled data by providing a means of leveraging unlabeled data. Since unlabeled data can often be obtained with minimal human labor, any performance boost conferred by SSL often comes with low cost. </p>\n\n<h3> Hence to win this challenge we invite you to read</h3>\n\n<ul>\n<li> https://amitness.com/2020/03/fixmatch-semi-supervised/ </li>\n<li> https://arxiv.org/abs/2001.07685 </li>\n<li>  https://arxiv.org/abs/1904.12848 </li>\n<li>https://arxiv.org/abs/1905.02249</li>\n<li>....</li>\n</ul>\n\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>The evaluation will be performed by evaluating your best model on the test set. \nPlease do not find the parameters of your best model on the test set. For this hyper parameter tuning task, use the validation set!</p>\n\n<p>Please evaluate the accuracy of the test set, and analyze your results.</p>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}