Time-series forecasting with deep learning: a survey

2021/02/15 by Bryan Lim, Stefan Zohren · 49 citations
Computer Science · Decision Sciences · #Forecasting Techniques and Applications #Stock Market Forecasting Methods #Time Series Analysis and Forecasting

paper · pdf · doi:10.1098/rsta.2020.0209

openalex publication_date 2021/02/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/31

Abstract

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'.

Citations

• Deep Gaussian Processes
• A tutorial on support vector regression
• Finding Structure in Time
• Neural Machine Translation by Jointly Learning to Align and Translate
• ImageNet classification with deep convolutional neural networks
• Representation Learning: A Review and New Perspectives
• Forecasting Sales by Exponentially Weighted Moving Averages
• Conditional Time Series Forecasting with Convolutional Neural Networks
• DeepAR: Probabilistic forecasting with autoregressive recurrent networks
• A Unified Approach to Interpreting Model Predictions
• Forecasting at Scale
• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• Conditional Neural Processes
• Neural Ordinary Differential Equations
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
• On a Method to Measure Supervised Multiclass Model’s Interpretability: Application to Degradation Diagnosis (Short Paper)
• RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes
• Exploiting Generative AI to Scale up Intelligent Tutoring Systems

Cited by

• Land surface information from satellites boost near-surface temperature forecast skill
• Signal decomposition–driven deep learning framework for passenger demand forecasting: Evidence from Dublin Luas
• Ergodic seismic precursors and transfer learning for short term eruption forecasting at data scarce volcanoes
• Time-Series Foundation Models for ISP Traffic Forecasting
• NetBurst: Event-Centric Forecasting of Bursty, Intermittent Time Series
• Unified Implementations of Recurrent Neural Networks in Multiple Deep Learning Frameworks
• Quantum Neural Network Architectures for Multivariate Time-Series Forecasting
• SEMPO: Lightweight Foundation Models for Time Series Forecasting
• Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency
• Online Kernel Dynamic Mode Decomposition for Streaming Time Series Forecasting with Adaptive Windowing
• Benchmarking Probabilistic Time Series Forecasting Models on Neural Activity
• LLM-as-a-Prophet: Understanding Predictive Intelligence with Prophet Arena
• Predicting the Unpredictable: Reproducible BiLSTM Forecasting of Incident Counts in the Global Terrorism Database (GTD)
• CrossAD: Time Series Anomaly Detection with Cross-scale Associations and Cross-window Modeling
• How Patterns Dictate Learnability in Sequential Data
• Variability Aware Recursive Neural Network (VARNN): A Residual-Memory Model for Capturing Temporal Deviation in Sequence Regression Modeling
• MLLM4TS: Leveraging Vision and Multimodal Language Models for General Time-Series Analysis
• MoGU: Mixture-of-Gaussians with Uncertainty-based Gating for Time Series Forecasting
• A Diffusion Model for Regular Time Series Generation from Irregular Data with Completion and Masking
• TimeFormer: Transformer with Attention Modulation Empowered by Temporal Characteristics for Time Series Forecasting
• Smart Contract Adoption under Discrete Overdispersed Demand: A Negative Binomial Optimization Perspective
• EntroPE: Entropy-Guided Dynamic Patch Encoder for Time Series Forecasting
• Echo Flow Networks
• Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting
• From Samples to Scenarios: A New Paradigm for Probabilistic Forecasting
• Information Bottleneck Learning for Faithful Time Series Forecasting Explanations
• Thinking in Categories: A Survey on Assessing the Quality for Time Series Synthesis
• Structural Forecasting for Tropical Cyclone Intensity Prediction: Providing Insight with Deep Learning
• DeltaHedge: A Multi-Agent Framework for Portfolio Options Optimization
• A Survey of Reasoning and Agentic Systems in Time Series with Large Language Models
• Short-Term Gaze Prediction: Analysis of Individual Differences, Typical and Extreme-Case Errors
• PLanTS: Periodicity-aware Latent-state Representation Learning for Multivariate Time Series
• AR-KAN: Autoregressive-Weight-Enhanced Kolmogorov-Arnold Network for Time Series Forecasting
• RDIT: Residual-based Diffusion Implicit Models for Probabilistic Time Series Forecasting
• A User-centric Kubernetes-based Architecture for Green Cloud Computing
• Quantum-Optimized Selective State Space Model for Efficient Time Series Prediction
• Enhancing Forecasting with a 2D Time Series Approach for Cohort-Based Data
• ELATE: Evolutionary Language model for Automated Time-series Engineering
• Deep Learning for Short Term Equity Trend Forecasting: A Behavior Driven Multi Factor Approach
• A Dual-Attention Graph Network for fMRI Data Classification
• Adaptive Fine-Tuning via Pattern Specialization for Deep Time Series Forecasting
• Stochastic dynamics learning with state-space systems
• EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecasting
• Salt-Rock Creep Deformation Forecasting Using Deep Neural Networks and Analytical Models for Subsurface Energy Storage Applications
• Long-Range Transformers for Dynamic Spatiotemporal Forecasting
• Towards Measuring and Modeling Geometric Structures in Time Series Forecasting via Image Modality
• Measuring Time-Series Dataset Similarity using Wasserstein Distance
• A Contrastive Diffusion-based Network (CDNet) for Time Series Classification
• Watermarking Large Language Model-based Time Series Forecasting
• From plots to commercial fields: scalable, transferable cotton morphology and productivity estimation using functional growth proxies from UAV and PlanetScope time series

Related

• A Survey of Deep Learning and Foundation Models for Time Series Forecasting
• What Matters in Deep Learning for Time Series Forecasting?
• Do We Really Need Deep Learning Models for Time Series Forecasting?
• A Comprehensive Survey of Deep Learning for Time Series Forecasting: Architectural Diversity and Open Challenges
• Time Series Analysis and Modeling to Forecast: a Survey