arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.04889math.OC
收藏

用于非光滑优化的投影近端梯度信赖域算法

Projected proximal gradient trust-region algorithm for nonsmooth optimization

Minh N. Dao, Hung M. Phan, Lindon Roberts

AI总结:

针对光滑非凸项与非光滑凸正则化项之和的优化问题,提出结合近端梯度下降与信赖域投影的子问题求解器,并建立无界 Hessian 增长下的复杂度与收敛保证。

摘要:

我们考虑用于求解目标函数为光滑非凸函数与非光滑凸正则化项之和的优化问题的信赖域方法。我们将此类方法的全局收敛理论扩展到模型 Hessian 增长无界情形下的最坏情况复杂度界,并提出一种新颖且简单的非光滑信赖域子问题求解器;该求解器结合若干次近端梯度下降迭代与一次到信赖域内的投影,满足算法收敛的充分下降要求,并取得了有前景的数值结果。

英文摘要:

We consider trust-region methods for solving optimization problems where the objective is the sum of a smooth, nonconvex function and a nonsmooth, convex regularizer. We extend the global convergence theory of such methods to include worst-case complexity bounds in the case of unbounded model Hessian growth, and introduce a new, simple nonsmooth trust-region subproblem solver based on combining several iterations of proximal gradient descent with a single projection into the trust region, which meets the sufficient descent requirements for algorithm convergence and has promising numerical results.

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。

↑