Delphi.science

Bias-Reduced and Parallel Heavy-Tailed Stochastic Convex Optimization in $\ell_p$ Geometries

Source: arxiv.org/abs/2611.21844

Zhengyun Nie, Anant Shyam, Brian Bullins

arXivarxiv/2611.21844Nov 16, 2026math.OCscore 0

We consider the fundamental problem of parallel stochastic convex optimization (SCO), whose aim is to approximately minimize a convex objective when given the ability to query a moment-bounded stochastic gradient oracle multiple times in parallel per round. Whereas most previous results have focused on the variance-bounded Euclidean ($\ell_2$-norm) setting, meaning the objective is Lipschitz with respect to the $\ell_2$-norm over an $\ell_2$-bounded feasible set in $\mathbb{R}^d$, in this work we study parallel SCO in $\ell_p$ settings for $2 \le p 0$ with at most $\tilde{O}(d^{\frac{p}{2p +…

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/2611.21844
curl -X POST /api/v1/records/arxiv/2611.21844/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.21844/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'