{
  "abstract": "Online learning in non-stationary streams is often formulated as tracking a point estimate, although many applications require predicting the full data-generating distribution. We study online distributional prediction under drift and bounded causal adversarial corruption. Candidate laws are represented by variable-size latent cluster configurations that induce predictive distributions. A Gibbs quasi-posterior over configurations yields an online predictor by posterior averaging and can be sampled using reversible-jump MCMC. The method uses no stream likelihood but specifies a kernel-mixture predictive family. We evaluate cumulative Wasserstein-1 regret against the time-varying true law and study periodic restart, sliding-window, and exponential-discounting memory against full history. The bounds separate PAC-Bayesian complexity, corruption-sensitive posterior perturbation, and drift through $A_T^{\\mathrm{OT}}=\\sum_{t=2}^T W_2^2(p_{t-1}^*,p_t^*)$. Under bounded support, latent quadratic growth, predictive-map regularity, and localized-comparator assumptions, restart has sublinear high-probability cumulative regret when $0\\le a<1$, $q,ζ,b<1$, $0\\leγ<β<1$, and $0<h<(1-a)/2$. Here $a,q,ζ,b$ control the polynomial growth of drift, predictive-family approximation error, off-order posterior mass, and oracle-order switch cost; $λ_s\\asymp s^{-β}$ and $H\\asymp T^h$. Corruptions satisfy the segment-local condition $Λ_{r,s}=O(s^γ)$, so a sublinear global budget alone is insufficient. The regime requires oracle-informed tuning, $\\varepsilon_T=o(1)$, and fixed confidence or $\\log(1/δ_0)=T^{o(1)}$. Sliding-window and discounted guarantees additionally require the stated localized-memory condition.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2606.18778",
  "author_total": 3,
  "authors": [
    {
      "name": "Navyansh Mahla"
    },
    {
      "name": "Prateek Chanda"
    },
    {
      "name": "Ganesh Ramakrishnan"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/exkvc3j8xn22s.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2606.18778",
  "fulltext_url": null,
  "id": "exkvc3j8xn22s",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2606.18778",
  "language": null,
  "licence": "CC-BY-NC-ND-4.0",
  "licence_label": "CC BY-NC-ND 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Data Stream Mining Techniques"
    ]
  ],
  "openalex": "W7165117798",
  "pdf_url": "https://arxiv.org/pdf/2606.18778",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-23",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/exkvc3j8xn22s.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:35:25Z",
      "native_id": "2606.18778",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2606.18778"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Online Distributional Prediction via Latent Cluster Geometry Under Drift and Corruption",
  "title_text": "Online Distributional Prediction via Latent Cluster Geometry Under Drift and Corruption",
  "type": "preprint",
  "updated_at": "2026-10-30T12:35:25Z",
  "url": "https://research.draconitic.com/paper/exkvc3j8xn22s",
  "venue": "arXiv",
  "versions": []
}