arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1809.02162cs.LGmath.OCstat.ML
收藏

约束最优化问题中鞍点的逃逸

Escaping Saddle Points in Constrained Optimization

Massachusetts Institute of Technology(麻省理工学院)

Aryan Mokhtari, Asuman Ozdaglar, Ali Jadbabaie

摘要:

本文研究了凸集约束下光滑非凸优化问题$\mathcal{C}$的鞍点逃逸问题。我们提出了一个通用的框架,产生收敛到二阶稳定点的问题,如果凸集$\mathcal{C}$是简单的二次目标函数。具体来说,我们的结果保持,如果可以找到一个$ρ$ -近似解的二次规划受$\mathcal{C}$在多项式时间,其中$ρ<1$是一个正常数,取决于结构的集合$\mathcal{C}$。在此条件下,我们表明,所提出的框架产生的迭代序列达到$(ε,γ)$ -二阶稳定点(SOSP)在最多$\mathcal{O}(\max\{ε^{-2},ρ^{-3}γ^{-3}\})$迭代。当凸集$\mathcal{C}$可以写成一组二次约束且目标函数Hessian在凸集$\mathcal{C}$上具有特定结构时,我们进一步刻画了达到SOSP的整体复杂性.最后,我们将我们的结果扩展到随机设置和随机梯度和Hessian评估的数量,以达到一个$(ε,γ)$ -SOSP。

英文摘要:

In this paper, we study the problem of escaping from saddle points in smooth nonconvex optimization problems subject to a convex set $\mathcal{C}$. We propose a generic framework that yields convergence to a second-order stationary point of the problem, if the convex set $\mathcal{C}$ is simple for a quadratic objective function. Specifically, our results hold if one can find a $ρ$-approximate solution of a quadratic program subject to $\mathcal{C}$ in polynomial time, where $ρ<1$ is a positive constant that depends on the structure of the set $\mathcal{C}$. Under this condition, we show that the sequence of iterates generated by the proposed framework reaches an $(ε,γ)$-second order stationary point (SOSP) in at most $\mathcal{O}(\max\{ε^{-2},ρ^{-3}γ^{-3}\})$ iterations. We further characterize the overall complexity of reaching an SOSP when the convex set $\mathcal{C}$ can be written as a set of quadratic constraints and the objective function Hessian has a specific structure over the convex set $\mathcal{C}$. Finally, we extend our results to the stochastic setting and characterize the number of stochastic gradient and Hessian evaluations to reach an $(ε,γ)$-SOSP.

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。

↑