{
  "abstract": "We propose a learning-to-optimize framework for solving parametric mixed-integer nonlinear programs arising in model predictive control and trajectory optimization. The proposed method combines neural network prediction with a sequential quadratic programming (SQP) refinement layer. First, neural networks are trained using a combination of supervised and self-supervised losses to predict integer solutions together with compatible continuous warm starts. To account for prediction inaccuracies, the trained predictor generates multiple candidate integer solutions, each paired with a continuous initialization. The integer solutions are then fixed, and a differentiable SQP layer refines the associated continuous variables over a prescribed number of iterations. The SQP-layer parameters, including the curvature regularization and step size, are trained to improve both feasibility and objective value after every iteration. By separating integer prediction from continuous optimization-based refinement, the proposed framework improves robustness to imperfect neural predictions while maintaining a predictable computational budget. We demonstrate the proposed framework on contact and trajectory optimization for a planar box-pushing example.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.05746",
  "author_total": 2,
  "authors": [
    {
      "name": "Viet-Anh Le"
    },
    {
      "name": "Rahul Mangharam"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5btc87cma3a7k.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.05746",
  "fulltext_url": null,
  "id": "5btc87cma3a7k",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.05746",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.SY"
    ],
    [
      "arxiv",
      "eess.SY"
    ],
    [
      "arxiv",
      "math.OC"
    ],
    [
      "openalex",
      "2207|Advanced Control Systems Optimization"
    ]
  ],
  "openalex": "W7220673270",
  "pdf_url": "https://arxiv.org/pdf/2610.05746",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-05",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/5btc87cma3a7k.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:10Z",
      "native_id": "2610.05746",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.05746"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:math.OC",
      "subfield": "optimisation-control"
    }
  ],
  "title": "Learning to Solve Mixed-Integer Nonlinear Programming with Neural Networks and Sequential Quadratic Programming Refinement",
  "title_text": "Learning to Solve Mixed-Integer Nonlinear Programming with Neural Networks and Sequential Quadratic Programming Refinement",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:10Z",
  "url": "https://research.draconitic.com/paper/5btc87cma3a7k",
  "venue": "arXiv",
  "versions": []
}