{
  "id": 417638,
  "title": "anyone can recommend transformer  + tabular synthetic data generation papers?",
  "url": "/competitions/predict-student-performance-from-game-play/discussion/417638",
  "author_name": "",
  "post_date": "2023-06-16T13:56:18.092608400Z",
  "votes": 7,
  "comment_count": 3,
  "views": 0,
  "content": "<p>i am a late comer to this competition.<br>\nI am interested in transformer method and tabular synthetic data generation (e.g. masked auto encoder or diffusion).</p>\n<p>anyone can recommend good papers?</p>\n<p>my plan:</p>\n<ol>\n<li>use neural net</li>\n<li>distill to catboost (regression of logic value)</li>\n</ol>",
  "messages": [
    {
      "id": "2305185",
      "postDate": "06/16/2023 13:56:18",
      "content": "<p>i am a late comer to this competition.<br>\nI am interested in transformer method and tabular synthetic data generation (e.g. masked auto encoder or diffusion).</p>\n<p>anyone can recommend good papers?</p>\n<p>my plan:</p>\n<ol>\n<li>use neural net</li>\n<li>distill to catboost (regression of logic value)</li>\n</ol>",
      "rawMarkdown": "i am a late comer to this competition.\nI am interested in transformer method and tabular synthetic data generation (e.g. masked auto encoder or diffusion).\n\nanyone can recommend good papers?\n\nmy plan:\n1. use neural net\n2. distill to catboost (regression of logic value)",
      "votes": null
    },
    {
      "id": "2305207",
      "postDate": "06/16/2023 14:09:35",
      "content": "<p>\"Datawig: Missing Data Imputation for Tabular Data with Deep Learning\" by Wang et al. (2019) - This paper introduces Datawig, a deep learning-based method for imputing missing values in tabular data. It utilizes a combination of deep learning techniques, including transformers, to effectively impute missing values and generate synthetic data.</p>\n<p>\"Transformer-based Synthetic Data Generation for Tabular Data\" by Alberici et al. (2021) - This paper proposes a method that combines transformers and generative models to generate synthetic data for tabular datasets. The transformer-based approach allows for capturing complex dependencies within the data, while the generative model component enables the generation of synthetic samples.</p>",
      "rawMarkdown": "\"Datawig: Missing Data Imputation for Tabular Data with Deep Learning\" by Wang et al. (2019) - This paper introduces Datawig, a deep learning-based method for imputing missing values in tabular data. It utilizes a combination of deep learning techniques, including transformers, to effectively impute missing values and generate synthetic data.\n\n\"Transformer-based Synthetic Data Generation for Tabular Data\" by Alberici et al. (2021) - This paper proposes a method that combines transformers and generative models to generate synthetic data for tabular datasets. The transformer-based approach allows for capturing complex dependencies within the data, while the generative model component enables the generation of synthetic samples.",
      "votes": null
    },
    {
      "id": "2306404",
      "postDate": "06/17/2023 09:20:48",
      "content": "<p><a href=\"https://github.com/Diyago/GAN-for-tabular-data\" target=\"_blank\">https://github.com/Diyago/GAN-for-tabular-data</a> <a href=\"https://arxiv.org/abs/2010.00638\" target=\"_blank\">https://arxiv.org/abs/2010.00638</a></p>",
      "rawMarkdown": "https://github.com/Diyago/GAN-for-tabular-data https://arxiv.org/abs/2010.00638",
      "votes": null
    },
    {
      "id": "2307291",
      "postDate": "06/18/2023 04:41:04",
      "content": "<p>This is great.. I am looking forward to such papers specially the first one. Thanks for sharing..🙏</p>",
      "rawMarkdown": "This is great.. I am looking forward to such papers specially the first one. Thanks for sharing..🙏",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2305207,
      "author_name": "ziancara",
      "author_url": "",
      "post_date": "06/16/2023 14:09:35",
      "content": "<p>\"Datawig: Missing Data Imputation for Tabular Data with Deep Learning\" by Wang et al. (2019) - This paper introduces Datawig, a deep learning-based method for imputing missing values in tabular data. It utilizes a combination of deep learning techniques, including transformers, to effectively impute missing values and generate synthetic data.</p>\n<p>\"Transformer-based Synthetic Data Generation for Tabular Data\" by Alberici et al. (2021) - This paper proposes a method that combines transformers and generative models to generate synthetic data for tabular datasets. The transformer-based approach allows for capturing complex dependencies within the data, while the generative model component enables the generation of synthetic samples.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2307291,
          "author_name": "cid007",
          "author_url": "",
          "post_date": "06/18/2023 04:41:04",
          "content": "<p>This is great.. I am looking forward to such papers specially the first one. Thanks for sharing..🙏</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2306404,
      "author_name": "insaff",
      "author_url": "",
      "post_date": "06/17/2023 09:20:48",
      "content": "<p><a href=\"https://github.com/Diyago/GAN-for-tabular-data\" target=\"_blank\">https://github.com/Diyago/GAN-for-tabular-data</a> <a href=\"https://arxiv.org/abs/2010.00638\" target=\"_blank\">https://arxiv.org/abs/2010.00638</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2305185": "i am a late comer to this competition.\nI am interested in transformer method and tabular synthetic data generation (e.g. masked auto encoder or diffusion).\n\nanyone can recommend good papers?\n\nmy plan:\n1. use neural net\n2. distill to catboost (regression of logic value)",
    "2305207": "\"Datawig: Missing Data Imputation for Tabular Data with Deep Learning\" by Wang et al. (2019) - This paper introduces Datawig, a deep learning-based method for imputing missing values in tabular data. It utilizes a combination of deep learning techniques, including transformers, to effectively impute missing values and generate synthetic data.\n\n\"Transformer-based Synthetic Data Generation for Tabular Data\" by Alberici et al. (2021) - This paper proposes a method that combines transformers and generative models to generate synthetic data for tabular datasets. The transformer-based approach allows for capturing complex dependencies within the data, while the generative model component enables the generation of synthetic samples.",
    "2306404": "https://github.com/Diyago/GAN-for-tabular-data https://arxiv.org/abs/2010.00638",
    "2307291": "This is great.. I am looking forward to such papers specially the first one. Thanks for sharing..🙏"
  },
  "source": "meta"
}