arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.02308math.OC
收藏

过参数化下鞍点问题的无梯度算法

Gradient-free algorithm for saddle point problems under overparametrization

Ekaterina Statkevich, Sofiya Bondar, Darina Dvinskikh, Alexander Gasnikov, Aleksandr Lobanov

AI总结:

本文研究过参数化且梯度计算不可行时的随机鞍点问题,通过推广Same-sample Stochastic Extra-gradient算法至有偏预言机,提出基于梯度近似的无梯度算法,并分析了对抗性噪声最大容许水平与最佳迭代次数。

摘要:

本文研究在过参数化机制下,当梯度计算不可行时,求解随机鞍点问题(SPP)。作为中间步骤,我们将Same-sample Stochastic Extra-gradient算法(Gorbunov et al., 2022)推广至有偏预言机,并估计了新的收敛速率。作为本文的结果,我们提出了一种使用梯度近似代替梯度预言机的算法。我们还进行了分析,以找出对抗性噪声的最大容许水平以及算法能保证达到所需精度的最佳迭代次数。

英文摘要:

This paper focuses on solving a stochastic saddle point problem (SPP) under an overparameterized regime for the case, when the gradient computation is impractical. As an intermediate step, we generalize Same-sample Stochastic Extra-gradient algorithm (Gorbunov et al., 2022) to a biased oracle and estimate novel convergence rates. As the result of the paper we introduce an algorithm, which uses gradient approximation instead of a gradient oracle. We also conduct an analysis to find the maximum admissible level of adversarial noise and the optimal number of iterations at which our algorithm can guarantee achieving the desired accuracy.

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。

↑