{
  "abstract": "Sparse variational Gaussian processes (SVGPs) are a widely used, principled approach to scalable Gaussian process modelling. Their success stems from their theoretically interpretable structure: they provide a lower bound on the marginal likelihood of the full GP while allowing scal- able stochastic optimization. Recent work has explored modifications of the posterior conditional distribution and has shown that such extensions can improve upon this well-established framework. In particular, introducing a block-diagonal structured scal- ing matrix leads to a tighter lower bound on the GP marginal likelihood. We investigate whether selecting blocks that tighten the variational objective improves test RMSE and NLL at a fixed inducing-point budget. Since the scaling matrix contributes to the variational lower bound through an upper bound on the log-determinant of the full residual covariance matrix, our work makes the following contributions. First, we introduce a simple, scalable, geometry-aware algorithm, with explicit computational complexity and low computational overhead, that recovers a large fraction of this log- determinant contribution by selecting an appropriate partition of the data. Second, we show how to train these models using a doubly stochastic optimization framework. Finally, we investigate how sparse GP methods based on alternative conditional posterior structures can be used to learn better hyperparameters and inducing points for SVGP.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.28256",
  "author_total": 1,
  "authors": [
    {
      "name": "Matteo Hoxha"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/eysmbfskekrdk.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.28256",
  "fulltext_url": null,
  "id": "eysmbfskekrdk",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.28256",
  "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",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Gaussian Processes and Bayesian Inference"
    ]
  ],
  "openalex": "W7223134029",
  "pdf_url": "https://arxiv.org/pdf/2610.28256",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/eysmbfskekrdk.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:34:13Z",
      "native_id": "2610.28256",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.28256"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Geometry-Aware Block Selection for Sparse Variational Gaussian Processes",
  "title_text": "Geometry-Aware Block Selection for Sparse Variational Gaussian Processes",
  "type": "preprint",
  "updated_at": "2026-10-30T12:34:13Z",
  "url": "https://research.draconitic.com/paper/eysmbfskekrdk",
  "venue": "arXiv",
  "versions": []
}