Closing the Accuracy Gap in Stochastic First-Order Bilevel Optimization

Haihan Zhang; Wendao Wu; Chenheng Zhang; Yanyi Li; Chunyuan Zheng; Cong Fang; Haoxuan Li; Zhouchen Lin

Closing the Accuracy Gap in Stochastic First-Order Bilevel Optimization

Haihan Zhang, Wendao Wu, Chenheng Zhang, Yanyi Li, Chunyuan Zheng, Cong Fang, Haoxuan Li, Zhouchen Lin

arXivarxiv/2610.03633Oct 2, 2026math.OCscore 0

We characterize the optimal accuracy dependence for smooth nonconvex--strongly-convex bilevel optimization with stochastic first-order oracles. For globally unbiased fresh-gradient observations with bounded variance, we prove an $Ω(ε^{-6})$ lower bound matching the accuracy exponent of existing upper bounds. We consider globally Lipschitz gradients, a bounded upper gradient in the lower variable, and Lipschitz lower Hessian blocks, with target $\mathbb{E}|\nabla F(\widehat{x})|\leε$. Let $L$ be the common first-order scale, $μ$ the lower strong-convexity modulus, $ρ$ the lower…

Source JSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2610.03633
curl -X POST /api/v1/records/arxiv/2610.03633/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.03633/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'