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 2209.04604math.OCcs.LG
收藏

凸-强凹鞍点问题的加速原始-对偶方法

Accelerated Primal-Dual Methods for Convex-Strongly-Concave Saddle Point Problems

Southern Methodist University(南卫理公会大学)

Mohammad Khalafi, Digvijay Boob

AI总结:

针对凸-强凹鞍点问题,提出单循环加速线性化原始-对偶方法(ALPD)及其非精确版本,通过结合加速梯度下降特性克服次优依赖,在半线性和一般非线性耦合下实现最优或显著改善的梯度复杂度。

摘要:

我们研究用于鞍点问题(SPP)的原始-对偶(PD)方法,该方法使用原始函数的线性近似代替标准的近端步骤,从而得到线性化PD(LPD)方法。对于凸-强凹SPP,我们观察到LPD方法对原始函数的Lipschitz常数具有次优依赖。为解决此问题,我们将加速梯度下降的特性与LPD方法相结合,提出一种单循环加速线性化原始-对偶(ALPD)方法。当SPP具有半线性耦合函数时,ALPD方法实现了最优的梯度复杂度。我们还针对具有一般非线性耦合函数的SPP提出了一种非精确ALPD方法,该方法保持了原始部分的最优梯度评估,并且与ALPD方法相比,显著改善了耦合项的梯度评估。我们通过数值实验验证了这些发现。

英文摘要:

We investigate a primal-dual (PD) method for the saddle point problem (SPP) that uses a linear approximation of the primal function instead of the standard proximal step, resulting in a linearized PD (LPD) method. For convex-strongly concave SPP, we observe that the LPD method has a suboptimal dependence on the Lipschitz constant of the primal function. To fix this issue, we combine features of Accelerated Gradient Descent with the LPD method resulting in a single-loop Accelerated Linearized Primal-Dual (ALPD) method. ALPD method achieves the optimal gradient complexity when the SPP has a semi-linear coupling function. We also present an inexact ALPD method for SPPs with a general nonlinear coupling function that maintains the optimal gradient evaluations of the primal parts and significantly improves the gradient evaluations of the coupling term compared to the ALPD method. We verify our findings with numerical experiments.

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。

↑