{
  "id": 16554914,
  "title": "Multiplicative vs. Additive State Coupling in Learned Optimizer Schedules",
  "url": "https://urgent.news/2026/10/23/multiplicative-vs-additive-state-coupling-in-learned-optimizer",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-23T15:40:23.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.32801v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "We learn schedules for a continuous penalty relaxation of MAX-CUT by backpropagating through an unrolled classical solver. A state-blind schedule generalizes to instances substantially larger than those used for training. A state-adaptive bilinear schedule also generalizes out of distribution, outperforming its untrained baseline across 15 seeds ($t=4.1$). By contrast, an additive ablation with…",
  "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."
}