{
  "abstract": "Deep randomized models fix hidden-layer parameters through random initialization and learn only closed-form readouts, typically adding depth by stacking random trans formations without target-aware control of hidden-state evolution. We propose LAIR Net, the Leaky Alignment-Impulse Residual Network, which mixes a shallow learned anchor into each hidden state through a leaky residual transition. We derive a depth uniform bound on input-perturbation sensitivity and use controlled simulations to attribute gains over a randomized baseline to the anchor rather than recursion or added capacity. Benefits emerge when a nonlinear target structure is learnable at the available noise level and diminish for nearly linear targets or dominant noise. Across 23 benchmark datasets, LAIR-Net achieves the best average rank among eight randomized networks and twelve conventional models, with relative performance associated with the same nonlinear-structure and noise quantities identified in simulation.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.11538",
  "author_total": 3,
  "authors": [
    {
      "name": "Rahul Goswami"
    },
    {
      "name": "Aryan Bhambu"
    },
    {
      "name": "Bittu Karmakar"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/1ebm51daqtatg.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.11538",
  "fulltext_url": null,
  "id": "1ebm51daqtatg",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.11538",
  "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",
      "1702|Neural Networks and Applications"
    ]
  ],
  "openalex": "W7221158533",
  "pdf_url": "https://arxiv.org/pdf/2610.11538",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-08",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/1ebm51daqtatg.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:28:03Z",
      "native_id": "2610.11538",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.11538"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "LAIR-Net: Leaky Alignment-Impulse Residual Networks for Tabular Regression",
  "title_text": "LAIR-Net: Leaky Alignment-Impulse Residual Networks for Tabular Regression",
  "type": "preprint",
  "updated_at": "2026-10-30T12:28:03Z",
  "url": "https://research.draconitic.com/paper/1ebm51daqtatg",
  "venue": "arXiv",
  "versions": []
}