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 1812.08306cs.LGstat.ML
收藏

NeuralWarp:基于Warping网络的时间序列相似性

NeuralWarp: Time-Series Similarity with Warping Networks

Information Systems and Machine Learning Lab(信息系统与机器学习实验室)

Josif Grabocka, Lars Schmidt-Thieme

摘要:

时间序列相似性度量的研究强调了弹性方法的必要性,这些方法可以调整时间序列对的索引,并且已经提出了大量的非参数方法。另一方面,深度学习方法在密切相关的领域占主导地位,例如学习图像和文本句子相似性。在本文中,我们提出了\textit{NeuralWarp},一种新的措施,模型对齐的时间序列索引在深表示空间,通过建模的翘曲函数作为一个上层神经网络之间的深度编码的时间序列值。实验结果表明,\textit{NeuralWarp}在一系列不同的现实生活数据集上的性能优于非参数和未扭曲的深度模型。

英文摘要:

Research on time-series similarity measures has emphasized the need for elastic methods which align the indices of pairs of time series and a plethora of non-parametric have been proposed for the task. On the other hand, deep learning approaches are dominant in closely related domains, such as learning image and text sentence similarity. In this paper, we propose \textit{NeuralWarp}, a novel measure that models the alignment of time-series indices in a deep representation space, by modeling a warping function as an upper level neural network between deeply-encoded time series values. Experimental results demonstrate that \textit{NeuralWarp} outperforms both non-parametric and un-warped deep models on a range of diverse real-life 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。

↑