Abstract:Deep learning, a crucial technique for achieving artificial intelligence (AI), has been successfully applied in many fields. The gradual application of the latest architectures of deep learning in the field of time series forecasting (TSF), such as Transformers, has shown excellent performance and results compared to traditional statistical methods. These applications are widely present in academia and in our daily lives, covering many areas including forecasting electricity consumption in power systems, meteorological rainfall, traffic flow, quantitative trading, risk control in finance, sales operations and price predictions for commercial companies, and pandemic prediction in the medical field. Deep learning-based TSF tasks stand out as one of the most valuable AI scenarios for research, playing an important role in explaining complex real-world phenomena. However, deep learning models still face challenges: they need to deal with the challenge of large-scale data in the information age, achieve longer forecasting ranges, reduce excessively high computational complexity, etc. Therefore, novel methods and more effective solutions are essential. In this paper, we review the latest developments in deep learning for TSF. We begin by introducing the recent development trends in the field of TSF and then propose a new taxonomy from the perspective of deep neural network models, comprehensively covering articles published over the past five years. We also organize commonly used experimental evaluation metrics and datasets. Finally, we point out current issues with the existing solutions and suggest promising future directions in the field of deep learning combined with TSF. This paper is the most comprehensive review related to TSF in recent years and will provide a detailed index for researchers in this field and those who are just starting out.

Neural Time Series Analysis with Fourier Transform: A Survey

A Survey on Deep Learning based Time Series Analysis with Frequency Transformation

QTFN: A General End-to-End Time-Frequency Network to Reveal the Time-Varying Signatures of the Time Series

Neural Fourier Modelling: A Highly Compact Approach to Time-Series Analysis

Deep Time Series Forecasting Models: A Comprehensive Survey

A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection

Recent Trends in Modelling the Continuous Time Series using Deep Learning: A Survey

Interpretable Multivariate Time Series Forecasting Using Neural Fourier Transform

Neural Architectures Learning Fourier Transforms, Signal Processing and Much More....

Time Series Analysis Based on Informer Algorithms: A Survey

FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective

Multivariate Time Series Anomaly Detection with Fourier Time Series Transformer

Deep Learning for Time Series Forecasting: A Survey

Transformers in Time Series: A Survey

FAN: Fourier Analysis Networks

Deep learning for time series classification: a review

New Neural Networks Based on Taylor Series and Their Research

Time Series Forecasting with Multi-scale Decomposition and Fourier Neural Operators

A novel hybrid deep learning time series forecasting model based on long-short-term patterns

A Comprehensive Survey of Time Series Forecasting: Architectural Diversity and Open Challenges