arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.03120cs.DScs.LGcs.NAmath.NA
收藏

面向序列组合最优传输的Sinkhorn算法

Sinkhorn Algorithm for Sequentially Composed Optimal Transports

National Institute of Informatics(信息学研究所);The University of Tokyo(东京大学)

Kazuki Watanabe, Noboru Isobe

AI总结:

针对序列组合最优传输的熵正则化问题,提出高效的Sinkhorn近似算法,并证明其在希尔伯特伪度量下的指数收敛性及单序列组合场景的最坏情况复杂度。

摘要:

Sinkhorn算法是最优传输领域事实上的标准近似算法,已应用于图像处理、自然语言处理等多种场景。理论上,其收敛性证明源于矩阵缩放问题的Sinkhorn--Knopp算法的收敛性,Altschuler等人证明其最坏情况时间复杂度为近线性时间。近期,Watanabe和Isobe提出了序列组合最优传输,作为最优传输的分层扩展。本文针对其熵正则化问题,提出了一种高效近似算法,即面向序列组合最优传输的Sinkhorn算法。此外,本文还对该Sinkhorn算法进行了理论分析:(i)在希尔伯特伪度量下,算法以指数速度收敛到最优解;(ii)针对单序列组合的情形,给出了最坏情况复杂度分析。

英文摘要:

Sinkhorn algorithm is the de-facto standard approximation algorithm for optimal transport, which has been applied to a variety of applications, including image processing and natural language processing. In theory, the proof of its convergence follows from the convergence of the Sinkhorn--Knopp algorithm for the matrix scaling problem, and Altschuler et al. show that its worst-case time complexity is in near-linear time. Very recently, sequentially composed optimal transports were proposed by Watanabe and Isobe as a hierarchical extension of optimal transports. In this paper, we present an efficient approximation algorithm, namely Sinkhorn algorithm for sequentially composed optimal transports, for its entropic regularization. Furthermore, we present a theoretical analysis of the Sinkhorn algorithm, namely (i) its exponential convergence to the optimal solution with respect to the Hilbert pseudometric, and (ii) a worst-case complexity analysis for the case of one sequential composition.

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。

↑