{
  "abstract": "Neural networks offer critical adaptability for controlling complex systems where nominal modeling proves intractable. However, when offline-trained neural networks are deployed within closed-loop control systems, their actions fundamentally alter the state distribution upon which they were trained. This paper introduces a unified taxonomy to categorize these sequential failure modes: the initial Closed-Loop Identification Error, the deployment-induced Action Error, and the continuous Tracking Error. Through discrete-time sensitivity analysis, this work formalizes how offline-learned parameters locally guarantee a trajectory divergence upon deployment, degrading performance due to local loss geometry. To mitigate this distribution shift, we propose actionable design policies centered on time-scale separated online adaptation. These mechanisms are validated through ablation studies on a nonlinear Duffing oscillator, demonstrating that bounded continuous learning successfully mitigates trajectory divergence whereas static deployments fail, offering concrete guidelines for practitioners deploying machine learning components in safety-critical feedback systems.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "Journal reference: (Lippi, 2026) In Proc. of the 23rd Int. Conf. on Informatics in Control, Automation and Robotics - Vol. 1, ISBN 978-989-758-870-9, ISSN 2184-2809, pp 15-26",
  "arxiv": "2610.22084",
  "author_total": 1,
  "authors": [
    {
      "name": "Vittorio Lippi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dcd1cdj0gfzv8.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.22084",
  "fulltext_url": null,
  "id": "dcd1cdj0gfzv8",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.22084",
  "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",
      "2207|Control Systems and Identification"
    ]
  ],
  "openalex": "W7222452338",
  "pdf_url": "https://arxiv.org/pdf/2610.22084",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-15",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dcd1cdj0gfzv8.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:30:00Z",
      "native_id": "2610.22084",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.22084"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Learning Changes the World and Then What Was Learned has Changed: Training and Deploying Neural Networks in the Control Loop",
  "title_text": "Learning Changes the World and Then What Was Learned has Changed: Training and Deploying Neural Networks in the Control Loop",
  "type": "preprint",
  "updated_at": "2026-11-04T15:20:53Z",
  "url": "https://research.draconitic.com/paper/dcd1cdj0gfzv8",
  "venue": "arXiv",
  "versions": [
    {
      "doi": "10.5220/0015198100004979",
      "pid": null,
      "relation": "published-as",
      "title": null,
      "venue": null
    }
  ]
}