Time-series forecasting with deep learning: a survey
Прогнозирование временных рядов с использованием глубокого обучения: обзор
2021-02-15
SCID: 54.1/9m93483q
Discuss with AI
deep learning architecturesencoder-decoder designshybrid deep learning modelsmulti-horizon forecastingtime-series forecasting
Figures from the paper
Abstract (AI)
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'.
Key Findings
1
Deep learning can support decision-making applications involving time-series data beyond forecasting alone.
2
Hybrid models combining established statistical methods with neural network components are highlighted as a way to improve purely statistical or purely neural approaches.
3
It explains how different forecasting models incorporate temporal information into their predictions across diverse time-series datasets and domains.
4
The survey categorizes common deep learning encoder and decoder architectures for one-step-ahead and multi-horizon time-series forecasting.
Research Object
deep learning architectures and hybrid models for time-series forecasting
Research Subject
encoder and decoder designs, temporal-information incorporation, hybrid model developments, and decision-support capabilities for one-step-ahead and multi-horizon time-series forecasting
Publication Details
Publication Date
2021-02-15
Journal
Publisher
ISSN
Cited by
1976
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai10
ImageNet classification with deep convolutional neural networks2017
Exploiting Generative AI to Scale up Intelligent Tutoring Systems2023
A Unified Approach to Interpreting Model Predictions2017
PyTorch: An Imperative Style, High-Performance Deep Learning Library2019
"Why Should I Trust You?": Explaining the Predictions of Any Classifier2016
Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation2014
Representation Learning: A Review and New Perspectives2013
Neural Machine Translation by Jointly Learning to Align and Translate2014
Transformer-XL: Attentive Language Models beyond a Fixed-Length Context2019
Qualitative Spatial Reasoning over Questions (Short Paper)2022
Cited by4
Transformers in Time Series: A Survey2023
A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection2024
Machine learning and deep learning based predictive quality in manufacturing: a systematic review2022
A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges2025