{
  "abstract": "Achieving local differential privacy in distributed optimization while maintaining low communication cost remains challenging. Existing vector quantization methods, such as vqSGD, rely on high-dimensional geometric constructions but incur unfavorable dimension-dependent variance. In this work, we propose Subsampled Stochastic TurboQuant (SSTQ), a framework that combines a bounded Kashin representation, data-independent coordinate subsampling, and privacy-aware one-dimensional quantization. SSTQ includes two variants: (1) a Flat Randomized Response variant that is unbiased and, for a fixed codebook bit-width, frame redundancy, and dimension-independent Kashin level, achieves reconstruction MSE that scales linearly with the ambient dimension $d$, while using only $\\lceil \\log_2 N \\rceil + b$ bits per message. Here, $N = Θ(d)$ denotes the frame size in the Kashin transform and $b$ is the codebook bit-width; and (2) a metric-aware truncated-Laplace variant that removes the exponential dependence on bit-width at the cost of a non-vanishing bias. We also derive a convex uniform-surrogate codebook objective whose worst-case codebook-dependent upper bound improves from $O(4^b)$ to $O(2^b)$. Experiments on synthetic regression, Fashion-MNIST, and CIFAR-10 compare the per-message privacy-utility and uplink-communication trade-offs of SSTQ with those of established baselines, demonstrating favorable utility and communication efficiency.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2608.05127",
  "author_total": 3,
  "authors": [
    {
      "name": "Adel Javanmard"
    },
    {
      "name": "David P. Woodruff"
    },
    {
      "name": "Vahab Mirrokni"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/8h9th22g383c3.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2608.05127",
  "fulltext_url": null,
  "id": "8h9th22g383c3",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2608.05127",
  "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|Privacy-Preserving Technologies in Data"
    ]
  ],
  "openalex": "W7202121614",
  "pdf_url": "https://arxiv.org/pdf/2608.05127",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-28",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/8h9th22g383c3.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:26Z",
      "native_id": "2608.05127",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2608.05127"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "SSTQ:Privacy-Preserving Vector Quantization via Subsampled Stochastic TurboQuant",
  "title_text": "SSTQ:Privacy-Preserving Vector Quantization via Subsampled Stochastic TurboQuant",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:26Z",
  "url": "https://research.draconitic.com/paper/8h9th22g383c3",
  "venue": "arXiv",
  "versions": []
}