{
  "fetched_at": "2026-10-20T03:07:07Z",
  "source_url": "https://www.kaggle.com/competitions/108-1-dl-app-hw3/data",
  "pages": [
    {
      "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
    },
    {
      "name": "data-description",
      "content": "Please download the VizDoom environment from <a href=\"https://github.com/mwydmuch/ViZDoom\">https://github.com/mwydmuch/ViZDoom</a>, and follow the instructions in \"HW3_vizdoom.pdf\".\n\n<table>\n<tr>\n<th>Filename </th><th> Description </th>\n</tr>\n<tr>\n<td>HW3_vizdoom.pdf </td><td> Homework 3 instructions</td>\n</tr>\n<tr>\n<td>ViZDoom on Colab.pdf </td><td> Training VizDoom on Colab</td>\n</tr>\n<tr>\n<td>D3_battle.wad </td><td> Training map</td>\n</tr>\n<tr>\n<td>D3_battle.cfg </td><td> Training configuration</td>\n</tr>\n<tr>\n<td>sample_ans.csv </td><td> Enter your best game score in this file, same score for both private and public datasets</td>\n</tr>\n</table>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p> In this homework, we are going to use reinforcement learning to train an AI warrior in Doom, a popular first-person shooting game first released in 1993. The VizDoom platform is created by Marek Wydmuch. Please visit his github:\n</p>\n\n<a>https://github.com/mwydmuch/ViZDoom</a>\n\n<p>\nWe've created our own training map \"D3_battle.wad\". The goal is to collect medial box while avoiding attacks from two monsters. Please enter your training scores manually in the sample CSV file and upload to Kaggle. We will randomly select some student models to verify.  After the competition, we will invite top 8 winners to have a death match, and let all players compete with each other in a new map.\n\n</p>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>\nIn this homework, you only need to upload your best test score. Since Kaggle requires at least two data rows, one for public and one for private data, you need to enter the same score twice. An example \"result.csv\" is shown below:\n</p>\n \n<table>\n<tbody>\n<tr>\n<th>Id</th><th>Predicted</th>\n</tr>\n<tr>\n<td>d3_battle_pubic</td><td><b>20</b></td>\n</tr>\n<tr>\n<td>d3_battle_private</td><td><b>20</b></td>\n</tr>\n</tbody></table>\n\n<p>\nPlease enter your game score (<b>20</b> in this example) into the two cells. However, we use Absolute Error (AE) as metric, and set the target as 100. So if you get a score 20 at \"D3_battle.wad\", you will get AE 80 on the leaderboad. Lower AE means better model. Kaggle will automatically convert your game score into AE. So the bottom line is: <b>Please Enter Your Game Score Directly!</b>.\n</p>\n<p></p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}