Delphi.science

Broximal Gradient Descent in General Norms: Convergence without Fejér Monotonicity

Source: arxiv.org/abs/2611.24645

Peter Richtárik, Xun Qian

arXivarxiv/2611.24645Nov 18, 2026math.OCscore 0

We study Broximal Gradient Descent in a general norm for smooth optimization over a closed convex set. Each update minimizes the linear model over the intersection of the feasible set and a norm ball. Unlike the Euclidean case with suitably restricted radii, the distance to a solution can increase, even when the squared norm is strongly convex. We instead use a comparator inequality and an observable stationarity certificate. For convex objectives, prescribed radii yield value convergence, including escape when the infimum is unattained, and a last-iterate $O(K^{-1/3})$ bound without a…

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/2611.24645
curl -X POST /api/v1/records/arxiv/2611.24645/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.24645/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'