{
  "abstract": "Generative modeling of spatio-temporal fields is crucial for a variety of applications, including stochastic weather generators and climate-model surrogates. However, many such fields exhibit complex dependence structures that vary across space and time and are nonlinear, resulting in nonstationary and non-Gaussian joint distributions. Our approach represents the joint density of a spatio-temporal field as a product of univariate conditional distributions and models these conditionals using Gaussian processes within an autoregressive transport-map construction. This prior distribution provides regularization, making our method suitable for a small number of training samples. Data-dependent sparsity in the conditioning sets ensures scalability to high-dimensional distributions. We also propose a variant of the method designed to sample or predict forward in time from a given incomplete space-time trajectory. We demonstrate the accuracy and scalability of our approach on non-Gaussian climate-model output with tens of millions of data points.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2605.03152",
  "author_total": 3,
  "authors": [
    {
      "name": "Carrie J. Lei-Cramer"
    },
    {
      "name": "Jian Cao"
    },
    {
      "name": "Matthias Katzfuss"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/c017pptmrtve0.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2605.03152",
  "fulltext_url": null,
  "id": "c017pptmrtve0",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2605.03152",
  "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",
      "stat.ME"
    ],
    [
      "openalex",
      "1702|Gaussian Processes and Bayesian Inference"
    ]
  ],
  "openalex": "W7160446379",
  "pdf_url": "https://arxiv.org/pdf/2605.03152",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-01",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/c017pptmrtve0.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:26:14Z",
      "native_id": "2605.03152",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2605.03152"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ME",
      "subfield": "statistics"
    }
  ],
  "title": "Scalable generative modeling of non-Gaussian spatio-temporal fields via autoregressive Gaussian processes",
  "title_text": "Scalable generative modeling of non-Gaussian spatio-temporal fields via autoregressive Gaussian processes",
  "type": "preprint",
  "updated_at": "2026-10-30T12:26:14Z",
  "url": "https://research.draconitic.com/paper/c017pptmrtve0",
  "venue": "arXiv",
  "versions": []
}