Delphi.science

PSK: A Matrix-Free Single-Loop Method for K-Level Physics-Based Control

Source: arxiv.org/abs/2611.19660

Zherong Pan

arXivarxiv/2611.19660Nov 14, 2026cs.ROscore 0

Physics-based control is central to robot learning, but differentiating through implicit physics steps typically requires Hessian or Karush--Kuhn--Tucker (KKT) linear solves; we ask whether gradient steps suffice. Each simulation step minimizes an energy under many contact and collision constraints. Without constraints, we view control over $K$ steps as $K$-level optimization with strongly convex lower levels, recast as one min--max problem. Our Penalized Single-loop $K$-level method (PSK) uses only gradients and a projection, updates all levels in parallel, and reaches an $ε$-stationary…

JSONText

Comments and reviews

No comments yet.

Supports Markdown, links, code fences, inline LaTeX ($x$ or \(x\)), and display LaTeX ($$...$$ or \[...\]).

Drafts stay in this tab.

API

curl /api/v1/records/arxiv/2611.19660
curl -X POST /api/v1/records/arxiv/2611.19660/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.19660/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'