{
  "abstract": "Machine-learning weather models (MLWMs) now match or outperform operational numerical weather prediction (NWP) at global medium-range forecasting, at far lower inference cost. Many deployed MLWMs are deterministic, producing a single forecast with no estimate of its own uncertainty, whereas a growing family of trained-probabilistic models generate calibrated ensembles directly, at the price of a dedicated training run. We ask instead how much uncertainty can be extracted from a deterministic checkpoint that already exists, without retraining it. Where physical ensembles represent model uncertainty by stochastically perturbing parametrisation tendencies, we perturb the network's raw weight tensors at inference time, a scheme we call stochastically perturbed weights (SPW). We also ask whether it works, where and on which scales to inject the noise, and where it fails. A three-phase ablation across four deterministic backbones, Aurora, GraphCast, SFNO, and AIFS, selects one production baseline per model, benchmarked against the trained-probabilistic AIFS-ENS, FourCastNet 3 and Atlas as well as the operational ECMWF ensemble (IFS-ENS) over 112 initialisation times. At a 240 h (10-day) lead time the SPW ensembles reach continuous ranked probability skill scores (CRPSS) between 0.04 and 0.13 below the best trained-probabilistic baseline, at zero marginal training cost. No injection site works across models: the productive tensor group is architecture-specific, so SPW is at present a tuning procedure rather than a plug-and-play recipe. Its main failure mode is a coherent whole-field offset that overdisperses the domain mean, and restricting the noise to coarse scales or perturbing the initial conditions each repair part of it.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.08412",
  "author_total": 4,
  "authors": [
    {
      "name": "Simon Adamov"
    },
    {
      "name": "Oliver Fuhrer"
    },
    {
      "name": "Reto Knutti"
    },
    {
      "name": "Sebastian Schemm"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/fdgrka16amg4q.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.08412",
  "fulltext_url": null,
  "id": "fdgrka16amg4q",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.08412",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "physics.ao-ph"
    ],
    [
      "openalex",
      "1902|Meteorological Phenomena and Simulations"
    ]
  ],
  "openalex": "W7212008703",
  "pdf_url": "https://arxiv.org/pdf/2609.08412",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-08",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/fdgrka16amg4q.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:19:49Z",
      "native_id": "2609.08412",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.08412"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.ao-ph",
      "subfield": "atmospheric"
    }
  ],
  "title": "Stochastically Perturbed Weights: Ensembles from Deterministic Machine-Learning Weather Models",
  "title_text": "Stochastically Perturbed Weights: Ensembles from Deterministic Machine-Learning Weather Models",
  "type": "preprint",
  "updated_at": "2026-10-30T12:19:49Z",
  "url": "https://research.draconitic.com/paper/fdgrka16amg4q",
  "venue": "arXiv",
  "versions": []
}