arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.23633cs.LG
收藏

单环随机双层优化的收敛性研究:近似隐式微分

On the Convergence of Single-Loop Stochastic Bilevel Optimization with Approximate Implicit Differentiation

Xi’an Jiaotong University(西安交通大学);Fudan University(复旦大学);SGIT AI Lab(SGIT人工智能实验室)

Yubo Zhou, Luo Luo, Guang Dai, Haishan Ye

AI总结:

本文提出单环随机近似隐式微分算法,证明其在O(κ⁷ ε⁻²)复杂度下达到ε-静态点,兼顾多环方法的最优收敛速度与单环的高效计算。

摘要:

随机双层优化已成为元学习和超参数优化的基本框架。尽管单环算法在实践中广泛使用,其理论理解,尤其是在随机环境下,仍远不如多环算法完善。现有分析往往得出次优收敛速度或模糊地将关键依赖于下层条件数κ的因素隐藏在通用Lipschitz常数中。在本文中,我们通过提供单环随机近似隐式微分(SSAID)算法的细化收敛分析来填补这一空白。我们证明SSAID在oracle复杂度下达到ε-静态点,复杂度为O(κ⁷ ε⁻²)。我们的结果有两点值得注意:(i) 它匹配了最先进的多环方法(如stocBiO)的最佳O(ε⁻²)速度,同时保持了单环更新的计算效率;(ii) 它为基于随机AID的单环方法提供了首次明确、细致的κ依赖性特征。这项工作证明SSAID不仅是一种启发式方法,而且具有严谨的理论基础,其收敛保证与主流多环框架相媲美。

英文摘要:

Stochastic Bilevel Optimization has emerged as a fundamental framework for meta-learning and hyperparameter optimization. Despite the practical prevalence of single-loop algorithms, their theoretical understanding in the stochastic regime remains less developed than that of multi-loop methods. In this paper, we provide a refined convergence analysis of the Single-loop Stochastic Approximate Implicit Differentiation (SSAID) algorithm. Under the squared-gradient stationarity criterion $\|\nablaΦ(x)\|^2\leε$, the corrected proof establishes an oracle complexity of $\mathcal{O}(κ^{14}ε^{-2})$, equivalently an averaged stationarity rate of $\mathcal{O}(κ^7K^{-1/2})$. The result preserves the canonical $\mathcal{O}(ε^{-2})$ dependence on the target accuracy while giving an explicit characterization of the condition-number dependence for stochastic AID-based single-loop methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑