arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.10065math.OC
收藏

又一种用于非凸优化的快速牛顿法变体

Yet another fast variant of Newton's method for nonconvex optimization

Serge Gratton, Sadok Jerad, Philippe L. Toint

AI总结:

提出一类在依赖迭代的子空间中交替正则化牛顿步与负曲率步的非凸优化二阶算法,并证明其一、二阶复杂度及数值有效性。

摘要:

本文提出了一类用于极小化光滑非凸函数的二阶算法,该算法在依赖迭代的子空间中交替采用正则化牛顿步和负曲率步。在大多数情况下,Hessian矩阵使用当前梯度的平方根以及一个考虑适度负曲率的附加项进行正则化,仅在例外情况下才采取负曲率步。文中详细介绍了若干实用变体,其中子空间被选为全空间或Krylov子空间。在第一种情况下,所提出的方法在几乎所有迭代中只需求解一个线性系统。我们证明,新类别中的算法获得ε近似一阶极小点至多需要O(|logε|ε^{-3/2})次问题目标函数及其导数求值,而获得二阶极小点至多需要O(|logε|ε^{-3})次求值。最后展示了针对两个全空间变体和两个Krylov子空间变体的令人鼓舞的初步数值实验。

英文摘要:

A class of second-order algorithms is proposed for minimizing smooth nonconvex functions that alternates between regularized Newton and negative curvature steps in an iteration-dependent subspace. In most cases, the Hessian matrix is regularized with the square root of the current gradient and an additional term taking moderate negative curvature into account, a negative curvature step being taken only exceptionally. Practical variants have been detailed where the subspaces are chosen to be the full space, or Krylov subspaces. In the first case, the proposed method only requires the solution of a single linear system at nearly all iterations. We establish that at most $\mathcal{O}\left(|\logε|\,ε^{-3/2}\right)$ evaluations of the problem's objective function and derivatives are needed for algorithms in the new class to obtain an $ε$-approximate first-order minimizer, and at most $\mathcal{O}\left(|\logε|\,ε^{-3}\right)$ to obtain a second-order one. Encouraging initial numerical experiments with two full-space and two Krylov-subspaces variants are finally presented.

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。

↑