{
  "abstract": "Stationarity rewards memory, but after a change the same history can mislead. We ask when forgetting should be permitted. E-process-authorized Thompson sampling (e-ATS) gives each arm full-history and discounted Beta states. An anytime-valid e-process first authorizes the discounted state, then a reversible relevance score controls its influence. Before authorization, e-ATS exactly follows optimistic Thompson sampling (OTS). Under a Beta-Bernoulli prior-predictive stationary model, e-ATS's probability of ever departing from OTS is at most the chosen $α_E$, without fitted thresholds. Relative to e-ATS, removing authorization increased mean normalized dynamic pseudo-regret by $38.4\\%$ on the registered suite but reduced it by $7.5\\%$ on the literature-derived replay suite. Therefore, evidence controls when adaptation begins, not whether it always helps.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.03646",
  "author_total": 3,
  "authors": [
    {
      "name": "Mayand Gulati"
    },
    {
      "name": "Kerong Wang"
    },
    {
      "name": "WeiChen Au"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/43rqpaqyy3r8a.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.03646",
  "fulltext_url": null,
  "id": "43rqpaqyy3r8a",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.03646",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1803|Advanced Bandit Algorithms Research"
    ]
  ],
  "openalex": "W7220451437",
  "pdf_url": "https://arxiv.org/pdf/2610.03646",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-02",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/43rqpaqyy3r8a.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:26:56Z",
      "native_id": "2610.03646",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.03646"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity",
  "title_text": "When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity",
  "type": "preprint",
  "updated_at": "2026-10-30T12:26:56Z",
  "url": "https://research.draconitic.com/paper/43rqpaqyy3r8a",
  "venue": "arXiv",
  "versions": []
}