{
  "id": 17176442,
  "title": "When to spend privacy: optimal allocation in private learning",
  "url": "https://urgent.news/2026/10/29/when-to-spend-privacy-optimal-allocation-in-private-learning",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-29T12:51:07.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.39954v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Privacy accounting bounds the cumulative privacy cost of Differentially Private Stochastic Gradient Descent but does not determine how to allocate a fixed budget across iterations. We formulate the complete Gaussian noise-variance schedule as an open-loop control under a composed zero-Concentrated Differential Privacy constraint, explicitly accounting for schedule selection. For expected…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}