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 2203.07844cs.LGcs.AI
收藏

哪种RNN单元结构最适合预测每种时间序列行为?

What is the best RNN-cell structure to forecast each time series behavior?

University of Granada(格拉纳达大学);ENSIAS, Mohammed V University of Rabat(拉巴特穆罕默德五世大学ENSIAS学院);Andalusian Research Institute in Data Science and Computational Intelligence, DaSCI(安达卢西亚数据科学与计算智能研究院)

Rohaifa Khaldi, Abdellatif El Afia, Raddouane Chiheb, Siham Tabik

AI总结:

本研究提出时间序列行为分类法,通过实验评估多种RNN单元结构,发现新变体SLIM在预测不同时间序列行为上表现优异且结构简单,计算成本低。

摘要:

毫无疑问,时间序列预测在许多领域都至关重要。解决时间序列预测任务最常用的机器学习模型是循环神经网络(RNNs)。通常,这些模型使用三种最流行的单元之一构建:ELMAN、长短期记忆(LSTM)或门控循环单元(GRU)。每种单元具有不同的结构,并意味着不同的计算成本。然而,目前尚不清楚为何以及何时使用每种RNN单元结构。实际上,没有对所有可能的时间序列行为进行全面表征,也没有指导哪种RNN单元结构最适合每种行为。本研究的目标有两个:提出一个几乎涵盖所有时间序列行为的全面分类法,并为每种时间序列行为提供最佳RNN单元结构的见解。我们进行了两个实验:(1)通过在其基本架构中创建一个变体(移除、添加或替换一个单元组件)来评估和分析LSTM-Vanilla单元中每个组件的作用,共创建11个变体。(2)评估和分析20种可能的RNN单元结构的性能。为了评估、比较和选择最佳模型,使用了不同的统计指标:基于误差的指标、基于信息准则的指标、基于朴素法的指标和基于方向变化的指标。为了进一步提高我们对模型解释和选择的信心,使用了Friedman Wilcoxon-Holm符号秩检验。我们的结果支持在时间序列预测中使用和探索新创建的RNN变体,名为SLIM,因为它具有高精度预测不同时间序列行为的能力,以及其简单的结构设计,不需要昂贵的时间和计算资源。

英文摘要:

It is unquestionable that time series forecasting is of paramount importance in many fields. The most used machine learning models to address time series forecasting tasks are Recurrent Neural Networks (RNNs). Typically, those models are built using one of the three most popular cells: ELMAN, Long Short-Term Memory (LSTM), or Gated Recurrent Unit (GRU) cells. Each cell has a different structure and implies a different computational cost. However, it is not clear why and when to use each RNN-cell structure. Actually, there is no comprehensive characterization of all the possible time series behaviors and no guidance on what RNN cell structure is the most suitable for each behavior. The objective of this study is twofold: it presents a comprehensive taxonomy of almost all time series behaviors and provides insights into the best RNN cell structure for each time series behavior. We conducted two experiments: (1) We evaluate and analyze the role of each component in the LSTM-Vanilla cell by creating 11 variants based on one alteration in its basic architecture (removing, adding, or substituting one cell component). (2) We evaluate and analyze the performance of 20 possible RNN-cell structures. To evaluate, compare, and select the best model, different statistical metrics were used: error-based metrics, information criterion-based metrics, naive-based metrics, and direction change-based metrics. To further improve our confidence in the models interpretation and selection, the Friedman Wilcoxon-Holm signed-rank test was used. Our results advocate the usage and exploration of the newly created RNN variant, named SLIM, in time series forecasting thanks to its high ability to accurately predict the different time series behaviors, as well as its simple structural design that does not require expensive temporal and computing resources.

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。

↑