论文标题
势头的复杂性保证polyak步骤
Complexity Guarantees for Polyak Steps with Momentum
论文作者
论文摘要
在平滑的强凸优化中，对强凸参数的了解对于获得加速速率的简单方法至关重要。在这项工作中，我们根据Polyak步骤来研究一类方法，其中这些知识被最佳值$ f _*$所代替。我们首先显示出比以前在使用Polyak步骤的简单梯度下降的经典情况下闻名的收敛边界的略有改善，然后我们得出了一种带有Polyak步骤和动量的加速梯度方法，以及收敛的保证。
In smooth strongly convex optimization, knowledge of the strong convexity parameter is critical for obtaining simple methods with accelerated rates. In this work, we study a class of methods, based on Polyak steps, where this knowledge is substituted by that of the optimal value, $f_*$. We first show slightly improved convergence bounds than previously known for the classical case of simple gradient descent with Polyak steps, we then derive an accelerated gradient method with Polyak steps and momentum, along with convergence guarantees.
