arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2307.04456math.OCcs.LG
收藏

Invex 规划:一阶算法及其收敛性

Invex Programs: First Order Algorithms and Their Convergence

Purdue University(普渡大学);Massachusetts Institute of Technology (MIT)(麻省理工学院)

Adarsh Barik, Suvrit Sra, Jean Honorio

AI总结:

本文针对经典一阶梯度方法求解 invex 规划收敛缓慢的问题,提出新的无约束一阶算法及首个带约束投影梯度方法,并给出收敛条件和速率保证。

摘要:

Invex 规划是一类特殊的非凸问题,其在每个驻点处都能达到全局最小值。经典的一阶梯度下降方法虽然可以求解这类问题,但收敛速度非常慢。本文提出了新的求解一般 invex 问题的一阶算法。我们确定了算法收敛的充分条件,并给出了收敛速率。此外,我们超越了无约束问题的范畴,针对带约束的 invex 规划提出了一种新颖的投影梯度方法,并提供了收敛速率保证。我们将所得结果与现有的一阶算法在多种无约束和带约束 invex 问题上进行了比较和对比。据我们所知,我们提出的算法是第一个能够求解带约束 invex 规划的算法。

英文摘要:

Invex programs are a special kind of non-convex problems which attain global minima at every stationary point. While classical first-order gradient descent methods can solve them, they converge very slowly. In this paper, we propose new first-order algorithms to solve the general class of invex problems. We identify sufficient conditions for convergence of our algorithms and provide rates of convergence. Furthermore, we go beyond unconstrained problems and provide a novel projected gradient method for constrained invex programs with convergence rate guarantees. We compare and contrast our results with existing first-order algorithms for a variety of unconstrained and constrained invex problems. To the best of our knowledge, our proposed algorithm is the first algorithm to solve constrained invex programs.

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。

↑