{
  "id": "2404.08168",
  "version": "v1",
  "published": "2024-04-12T00:21:30.000Z",
  "updated": "2024-04-12T00:21:30.000Z",
  "title": "Conformal Prediction via Regression-as-Classification",
  "authors": [
    "Etash Guha",
    "Shlok Natarajan",
    "Thomas Möllenhoff",
    "Mohammad Emtiyaz Khan",
    "Eugene Ndiaye"
  ],
  "comment": "International Conference of Learning Representations 2024",
  "journal": "International Conference of Learning Representations 2024",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "Conformal prediction (CP) for regression can be challenging, especially when the output distribution is heteroscedastic, multimodal, or skewed. Some of the issues can be addressed by estimating a distribution over the output, but in reality, such approaches can be sensitive to estimation error and yield unstable intervals.~Here, we circumvent the challenges by converting regression to a classification problem and then use CP for classification to obtain CP sets for regression.~To preserve the ordering of the continuous-output space, we design a new loss function and make necessary modifications to the CP classification techniques.~Empirical results on many benchmarks shows that this simple approach gives surprisingly good results on many practical problems.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2024-04-12T00:21:30.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "conformal prediction",
      "regression-as-classification",
      "output distribution",
      "simple approach",
      "cp classification"
    ],
    "tags": [
      "conference paper",
      "journal article"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}