{
  "annotation_id": "933887a9-de72-4dc0-b91c-696af20c1af6",
  "date_created": "2026-02-17T05:53:23.979000Z",
  "date_modified": "2026-02-17T05:53:23.979000Z",
  "file_hash": "96369b26f4a28f2d275b3886849cdf1a0ba89bbb9965f9e09bcbb07b76021e49",
  "private": false,
  "record": {
    "abstract": "Dimension reduction is a fundamental tool for analyzing high-dimensional data in supervised learning. Traditional methods for estimating intrinsic order often prioritize model-specific structural assumptions over predictive utility. This paper introduces predictive order determination (POD), a model-agnostic framework that determines the minimal predictively sufficient dimension by directly evaluating out-of-sample predictiveness. POD quantifies uncertainty via error bounds for over- and underestimation and achieves consistency under mild conditions. By unifying dimension reduction with predictive performance, POD applies flexibly across diverse reduction tasks and supervised learners. Simulations and real-data analyses show that POD delivers accurate, uncertainty-aware order estimates, making it a versatile component for prediction-centric pipelines.",
    "arxiv_id": "2601.10357",
    "authors": [
      "Yue Yu",
      "Guanghui Wang",
      "Liu Liu",
      "Changliang Zou"
    ],
    "categories": [
      "stat.ME",
      "math.ST",
      "stat.TH"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "Model-Agnostic and Uncertainty-Aware Dimensionality Reduction in Supervised Learning",
    "url": "https://arxiv.org/abs/2601.10357",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "02364996-8186-4487-86b9-9134cec94930",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}