arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.12751math.OCcs.LG
收藏

快速随机复合最小化与并行化下的加速Frank-Wolfe算法

Fast Stochastic Composite Minimization and an Accelerated Frank-Wolfe Algorithm under Parallelization

Ecole normale supérieure(巴黎高等师范学院);Université PSL(巴黎文理研究大学);CNRS(法国国家科学研究中心);INRIA(法国国家信息与自动化研究所);Google Research(谷歌研究院)

Benjamin Dubois-Taine, Francis Bach, Quentin Berthet, Adrien Taylor

AI总结:

本文针对随机复合凸优化问题提出一种Bregman型加速算法,并基于该框架设计出在并行化下实现加速的Frank-Wolfe变体,以O(1/√ε)次迭代达到ε原始-对偶间隙。

摘要:

我们考虑最小化两个凸函数之和的问题。其中一个函数具有Lipschitz连续梯度,并且可以通过随机oracle访问,而另一个函数是“简单的”。我们提供了一种Bregman型算法,其函数值以加速收敛到一个包含最小值的球。该球的半径取决于与问题相关的常数,包括随机oracle的方差。我们进一步表明,这种算法设置自然地引出了Frank-Wolfe的一种变体,在并行化下实现加速。更准确地说,当在有界域上最小化光滑凸函数时,我们表明,通过仅访问原始函数的梯度以及一个线性最大化oracle,并使用O(1/√ε)个并行计算单元,可以在O(1/√ε)次迭代内(期望意义上)达到ε原始-对偶间隙。我们在合成数值实验上展示了这种快速收敛性。

英文摘要:

We consider the problem of minimizing the sum of two convex functions. One of those functions has Lipschitz-continuous gradients, and can be accessed via stochastic oracles, whereas the other is "simple". We provide a Bregman-type algorithm with accelerated convergence in function values to a ball containing the minimum. The radius of this ball depends on problem-dependent constants, including the variance of the stochastic oracle. We further show that this algorithmic setup naturally leads to a variant of Frank-Wolfe achieving acceleration under parallelization. More precisely, when minimizing a smooth convex function on a bounded domain, we show that one can achieve an $ε$ primal-dual gap (in expectation) in $\tilde{O}(1/ \sqrtε)$ iterations, by only accessing gradients of the original function and a linear maximization oracle with $O(1/\sqrtε)$ computing units in parallel. We illustrate this fast convergence on synthetic 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。

↑