{
  "annotation_id": "a86cbe91-5f62-4e54-bb6e-f2cb6600c4bb",
  "date_created": "2026-03-02T18:01:32.328000Z",
  "date_modified": "2026-03-02T18:01:32.328000Z",
  "file_hash": "4593775d6f90c75731d380ba64244418dfd418e084b47215b96ef19f388da477",
  "private": false,
  "record": {
    "abstract": "Advances in statistical learning theory have resulted in a multitude of\ndifferent designs of learning machines. But which ones are implemented by\nbrains and other biological information processors? We analyze how various\nabstract Bayesian learners perform on different data and argue that it is\ndifficult to determine which learning-theoretic computation is performed by a\nparticular organism using just its performance in learning a stationary target\n(learning curve). Basing on the fluctuation-dissipation relation in statistical\nphysics, we then discuss a different experimental setup that might be able to\nsolve the problem.",
    "arxiv_id": "q-bio/0402029",
    "authors": [
      "Ilya Nemenman"
    ],
    "categories": [
      "q-bio.NC",
      "cs.LG",
      "nlin.AO",
      "physics.data-an"
    ],
    "journal_ref": "Neural Comp. 17 (9): 2006-2033 SEP 2005",
    "title": "Fluctuation-dissipation theorem and models of learning",
    "url": "https://arxiv.org/abs/q-bio/0402029"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "c4f896a8-cd6a-4b22-8855-65e5bb8ad41e",
    "id": "arXiv Dataset IDs",
    "type": "Model",
    "variant": "snapshot-2026-03-01",
    "version": "0.1.0"
  },
  "user_id": 1000002
}