arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.09222cs.LG
收藏

多尺度流形翘曲

Multiscale Manifold Warping

Adobe Research(奥多比研究院)

Sridhar Mahadevan, Anup Rao, Georgios Theocharous, Jennifer Healey

摘要:

许多现实世界的应用需要对齐两个时间序列,包括生物信息学,手写识别,活动识别和人机协调。动态时间规整(DTW)是一种流行的比对方法,但在高维真实世界的数据中可能会失败,其中比对序列的维度通常是不相等的。在本文中,我们表明,利用现实世界的数据的多尺度流形的潜在结构,可以产生改进的对齐。我们介绍了一种新的框架,称为Warping on Wavelets(WOW),它将DTW与称为Diffusion Wavelets的多尺度流形学习框架相结合。我们提出了一个理论分析的WOW家族的算法,并表明它优于以前的最先进的方法,如规范时间规整(CTW)和流形翘曲,在几个现实世界的数据集。

英文摘要:

Many real-world applications require aligning two temporal sequences, including bioinformatics, handwriting recognition, activity recognition, and human-robot coordination. Dynamic Time Warping (DTW) is a popular alignment method, but can fail on high-dimensional real-world data where the dimensions of aligned sequences are often unequal. In this paper, we show that exploiting the multiscale manifold latent structure of real-world data can yield improved alignment. We introduce a novel framework called Warping on Wavelets (WOW) that integrates DTW with a a multi-scale manifold learning framework called Diffusion Wavelets. We present a theoretical analysis of the WOW family of algorithms and show that it outperforms previous state of the art methods, such as canonical time warping (CTW) and manifold warping, on several real-world datasets.

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。

↑