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 2305.12957math.OC
收藏

分布式在线多重 Frank-Wolfe 凸优化的改进动态遗憾

Improved Dynamic Regret of Distributed Online Multiple Frank-Wolfe Convex Optimization

Wentao Zhang, Yang Shi, Baoyong Zhang, Deming Yuan

AI总结:

针对时变多智能体网络中的高维约束在线凸优化,提出分布式在线多重 Frank-Wolfe 算法,免投影并改进动态遗憾界,揭示计算与通信权衡。

摘要:

本文考虑时变多智能体网络上的分布式在线凸优化问题。该网络的目标是通过局部计算以及与邻居通信来最小化全局损失函数。为有效处理具有高维结构化约束集的优化问题,我们开发了一种分布式在线多重 Frank-Wolfe 算法,以避免投影操作带来的高昂计算代价。所建立的动态遗憾界为 O(T^{1-γ}+H_T),线性预言机调用次数为 O(T^{1+γ}),其中该界取决于时域(总迭代次数)T、函数变化量 H_T 以及调节参数 0<γ<1。特别地,当能够获得 H_T 和 T 的先验知识时,该界可增强为 O(1+H_T)。此外,我们阐明了多重迭代技术的显著优势,并揭示了动态遗憾界、计算成本与通信成本之间的权衡。最后,通过具有两个约束集的分布式在线岭回归问题,验证并比较了我们算法的性能。

英文摘要:

In this paper, we consider a distributed online convex optimization problem over a time-varying multi-agent network. The goal of this network is to minimize a global loss function through local computation and communication with neighbors. To effectively handle the optimization problem with a high-dimensional and structural constraint set, we develop a distributed online multiple Frank-Wolfe algorithm to avoid the expensive computational cost of projection operation. The dynamic regret bounds are established as $\mathcal{O}(T^{1-γ}+H_T)$ with the linear oracle number $\mathcal{O} (T^{1+γ})$, which depends on the horizon (total iteration number) $T$, the function variation $H_T$, and the tuning parameter $0<γ<1$. In particular, when the prior knowledge of $H_T$ and $T$ is available, the bound can be enhanced to $\mathcal{O} (1+H_T)$. Moreover, we illustrate the significant advantages of the multiple iteration technique and reveal a trade-off between dynamic regret bound, computational cost, and communication cost. Finally, the performance of our algorithm is verified and compared through the distributed online ridge regression problems with two constraint sets.

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。

↑