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 1906.03056math.OC
收藏

自适应快速梯度法的Polyak步长

Polyak Steps for Adaptive Fast Gradient Method

Mathieu Barré, Alexandre d'Aspremont

摘要:

光滑强凸函数极小化的加速算法通常需要强凸性参数$μ$的知识。在未知的$μ$的情况下,当前的自适应技术基于重启方案。当最优值$f^*$已知时,这些策略恢复加速的线性收敛界而无需额外的网格搜索。在本文中,我们提出了一种新的方法,具有相同的界没有任何重新启动,使用在线估计的强凸性参数。我们显示的快速梯度法的鲁棒性时,使用的序列上界$μ$。我们还提出了一个很好的候选人,这个估计序列和细节一致的实证结果。

英文摘要:

Accelerated algorithms for minimizing smooth strongly convex functions usually require knowledge of the strong convexity parameter $μ$. In the case of an unknown $μ$, current adaptive techniques are based on restart schemes. When the optimal value $f^*$ is known, these strategies recover the accelerated linear convergence bound without additional grid search. In this paper we propose a new approach that has the same bound without any restart, using an online estimation of strong convexity parameter. We show the robustness of the Fast Gradient Method when using a sequence of upper bounds on $μ$. We also present a good candidate for this estimate sequence and detail consistent empirical 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。

↑