论文标题

在小批次培训中,长度时间序列不同

On Mini-Batch Training with Varying Length Time Series

论文作者

Iwana, Brian Kenji

论文摘要

在现实世界的时间序列识别应用程序中,可以拥有具有不同长度模式的数据。但是,当使用人工神经网络(ANN)时,使用固定尺寸的迷你批次是标准做法。为此,通常将长度变化的时间序列数据归一化,因此所有模式的长度相同。通常,这是使用零填充或截断完成的,而无需考虑太多。我们提出了一种新颖的方法,可以利用动态时间扭曲(DTW)的动态匹配能力来规范数据集中的时间序列。这样,数据集中的时间序列长度可以设置为固定大小,同时维护数据集典型的功能。在实验中,使用了2018年UCR时间序列档案中的所有11个具有不同长度时间序列的数据集。我们通过将其与卷积神经网络(CNN),长期术语内存网络(LSTM)和双向LSTM(BLSTM)上的18种其他长度归一化方法进行比较来评估所提出的方法。

In real-world time series recognition applications, it is possible to have data with varying length patterns. However, when using artificial neural networks (ANN), it is standard practice to use fixed-sized mini-batches. To do this, time series data with varying lengths are typically normalized so that all the patterns are the same length. Normally, this is done using zero padding or truncation without much consideration. We propose a novel method of normalizing the lengths of the time series in a dataset by exploiting the dynamic matching ability of Dynamic Time Warping (DTW). In this way, the time series lengths in a dataset can be set to a fixed size while maintaining features typical to the dataset. In the experiments, all 11 datasets with varying length time series from the 2018 UCR Time Series Archive are used. We evaluate the proposed method by comparing it with 18 other length normalization methods on a Convolutional Neural Network (CNN), a Long-Short Term Memory network (LSTM), and a Bidirectional LSTM (BLSTM).

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源