arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.08618math.OC
收藏

Nesterov加速梯度法用于无界凸函数在对偶空间中寻找最小范数点

Nesterov's accelerated gradient for unbounded convex functions finds the minimum-norm point in the dual space

Faculty of Computer Science, Ruhr University Bochum(鲁尔大学波鸿分校计算机科学学院)

Keiya Sakabe

AI总结:

本文研究了Nesterov加速梯度法在无界凸函数中寻找最小范数点的收敛性,揭示了其在对偶空间中的应用及加速效果。

摘要:

我们研究了一阶方法应用于下界无界的凸函数$f$的行为,即$\inf f = -\infty$。这种设置由于梯度下降和Nesterov加速梯度方法的轨迹发散而受到很少关注。在本文中,我们建立了定量收敛结果,描述了它们的速度和发散方向,对无界性的判断有启示。关键思想是与对偶空间中的一个范数最小化问题的关系:最小化$\|p\|^2/2$ over $p \in \mathrm{dom}f^\ast$,这可以通过取Legendre-Fenchel共轭$f^\ast$作为距离生成函数来自然地通过镜像下降法解决。然后发现,对于$f$的梯度下降法与该范数最小化问题的镜像下降法相同,因此在$\mathcal{O}(k^{-1})$范围内同时解决两个问题。该结果允许加速;Nesterov加速梯度法在不进行任何修改的情况下,同时在$\mathcal{O}(k^{-2})$范围内解决原始最小化和对偶范数最小化问题,为无界凸优化中的发散提供了定量描述。

英文摘要:

We study the behavior of first-order methods applied to a lower-unbounded convex function $f$, i.e., $\inf f = -\infty$. Such a setting has received little attention since the trajectories of gradient descent and Nesterov's accelerated gradient method diverge. In this paper, we establish quantitative convergence results describing their speeds and directions of divergence, with implications for unboundedness judgment. A key idea is a relation to a norm-minimization problem in the dual space: minimize $\|p\|^2/2$ over $p \in \mathrm{dom}f^\ast$, which can be naturally solved via mirror descent by taking the Legendre--Fenchel conjugate $f^\ast$ as the distance-generating function. It then turns out that gradient descent for $f$ coincides with mirror descent for this norm-minimization problem, and thus it simultaneously solves both problems at $\mathcal{O}(k^{-1})$. This result admits acceleration; Nesterov's accelerated gradient method, without any modifications, simultaneously solves the original minimization and the dual norm-minimization problems at $\mathcal{O}(k^{-2})$, providing a quantitative characterization of divergence in unbounded convex optimization.

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。

↑