{
  "abstract": "Large-scale linear programs with millions of variables and constraints strain time and memory budgets. First-order methods such as the primal-dual hybrid gradient scale to this regime because they avoid matrix factorizations, but they require many iterations. Learned accelerators can reduce these iterations, yet fewer iterations do not guarantee a shorter solve. Learning inference also takes time, and running the accelerator longer trades solver time for inference time. Existing unrolled accelerators handle this trade-off poorly: each layer processes every variable and constraint, the predicted iterates still need solver refinement, and every instance runs the same layers. We therefore ask: What neural accelerator design achieves end-to-end speedups on large-scale linear programs across sizes and target accuracies? We introduce AHEAD, an accelerator that predicts step sizes for first-order linear programming solvers while keeping their update rules. Its weight-shared predictor reads compact solver statistics and rescales the primal and dual step sizes, so each warm-start iteration stays inexpensive and a short warm-start phase hands the solver a warm start. A learned handoff policy sets the number of warm-start iterations for each instance and target accuracy, after which the solver finishes under its own termination criterion. On packing linear programs with up to $10^7$ variables, AHEAD achieves $6.5$--$11.6\\times$ end-to-end speedups over Gurobi's first-order solver from a cold start, including on instances of unseen sizes and distributions. The gains extend to PageRank linear programs and hold across difficulty levels and target accuracies.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.36321",
  "author_total": 7,
  "authors": [
    {
      "name": "Sihao Liu"
    },
    {
      "name": "Zhichao Hou"
    },
    {
      "name": "Huayue Gu"
    },
    {
      "name": "Raiyan Rahman"
    },
    {
      "name": "Liang Wu"
    },
    {
      "name": "Ruozhou Yu"
    },
    {
      "name": "Xiaorui Liu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/c7s4e3gwj26m6.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.36321",
  "fulltext_url": null,
  "id": "c7s4e3gwj26m6",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.36321",
  "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",
      "math.OC"
    ],
    [
      "openalex",
      "2612|Advanced Optimization Algorithms Research"
    ]
  ],
  "openalex": "W7224094757",
  "pdf_url": "https://arxiv.org/pdf/2610.36321",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-26",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/c7s4e3gwj26m6.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:35:53Z",
      "native_id": "2610.36321",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.36321"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:math.OC",
      "subfield": "optimisation-control"
    }
  ],
  "title": "AHEAD: Adaptive Neural Acceleration for Large-Scale Linear Programming",
  "title_text": "AHEAD: Adaptive Neural Acceleration for Large-Scale Linear Programming",
  "type": "preprint",
  "updated_at": "2026-10-30T12:35:53Z",
  "url": "https://research.draconitic.com/paper/c7s4e3gwj26m6",
  "venue": "arXiv",
  "versions": []
}