Oxford-Man Institute for Quantitative Finance, Department of Engineering Science, University of Oxford, Oxford, UK.
Philos Trans A Math Phys Eng Sci. 2021 Apr 5;379(2194):20200209. doi: 10.1098/rsta.2020.0209. Epub 2021 Feb 15.
Numerous deep learning architectures have been developed to accommodate the diversity of time-series datasets across different domains. In this article, we survey common encoder and decoder designs used in both one-step-ahead and multi-horizon time-series forecasting-describing how temporal information is incorporated into predictions by each model. Next, we highlight recent developments in hybrid deep learning models, which combine well-studied statistical models with neural network components to improve pure methods in either category. Lastly, we outline some ways in which deep learning can also facilitate decision support with time-series data. This article is part of the theme issue 'Machine learning for weather and climate modelling'.
已经开发出许多深度学习架构来适应不同领域的时间序列数据集的多样性。在本文中，我们调查了在单步和多步时间序列预测中使用的常见编码器和解码器设计，描述了每个模型如何将时间信息纳入预测。接下来，我们重点介绍了混合深度学习模型的最新进展，这些模型将经过充分研究的统计模型与神经网络组件相结合，以改进这两类方法中的纯方法。最后，我们概述了深度学习如何利用时间序列数据来促进决策支持。本文是“机器学习在天气和气候建模中的应用”主题特刊的一部分。
