{
  "abstract": "We propose a method for choosing the shared memory parameter $β_1=β_2=β$ in Adam from a short pilot training. The selected $β$ remains fixed during the subsequent full training. A local model of Adam's normalized direction balances sampling variability against the delay introduced by averaging past gradients. This balance gives a cubic memory rule, whose two coefficients are estimated from gradient probes at a few pilot checkpoints. The estimator uses the numerator and denominator jointly, preserving their covariance. With a 200-update pilot and sixteen probe gradients at each of four checkpoints, a seed-matched retrospective evaluation on eleven vision and language workloads reduces mean relative validation gap by 40.7% and worst-quarter mean gap by 44.3% against the grid representative of shared $β=0.95$. The mean gap is also 32.3% lower than that of the best constant $β$ chosen across all eleven workloads.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.08624",
  "author_total": 5,
  "authors": [
    {
      "name": "Alberto Fernández-Hernández"
    },
    {
      "name": "Cristian Pérez-Corral"
    },
    {
      "name": "Jose I. Mestre"
    },
    {
      "name": "Manuel F. Dolz"
    },
    {
      "name": "Enrique S. Quintana-Ortí"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/f13h43t85tk0q.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.08624",
  "fulltext_url": null,
  "id": "f13h43t85tk0q",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.08624",
  "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.AI"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Stochastic Gradient Optimization Techniques"
    ]
  ],
  "openalex": "W7221054963",
  "pdf_url": "https://arxiv.org/pdf/2610.08624",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-06",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/f13h43t85tk0q.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:32Z",
      "native_id": "2610.08624",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.08624"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Early Memory Selection for Balanced Adam",
  "title_text": "Early Memory Selection for Balanced Adam",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:32Z",
  "url": "https://research.draconitic.com/paper/f13h43t85tk0q",
  "venue": "arXiv",
  "versions": []
}