Foundations of Sequence-to-Sequence Modeling for Time Series
Explore this paper's citation graph
Summary
The first theoretical analysis of this time series forecasting framework is provided, including a comparison of sequence-to-sequence modeling to classical time series models that can serve as a quantitative guide for practitioners choosing between different modeling methodologies.
- Type
- preprint
- Published
- 2018-05-09
- Cited by
- 66
- References
- 45
- Access
- Open access
- OpenAlex
- https://openalex.org/W2800296055
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:13673179
Keywords
Sequence (biology), Series (stratigraphy), Computer science, Time sequence, Time series
References
- Climate Prediction via Matrix Completion
- Mixing: Properties and Examples
- Empirical margin distributions and bounding the generalization error of combined classifiers
- Global Climate Model Tracking Using Geospatial Neighborhoods
- Regularized estimation in sparse high-dimensional time series models
- Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation
- Sequential complexities and uniform martingale laws of large numbers
- Time series analysis, forecasting and control
- Traffic Flow Prediction With Big Data: A Deep Learning Approach
- Large Bayesian vector auto regressions
- Long Short-Term Memory
- RATES OF CONVERGENCE FOR EMPIRICAL PROCESSES OF STATIONARY MIXING SEQUENCES
- Nonparametric Time Series Prediction Through Adaptive Model Selection
- Estimation of (near) low-rank matrices with noise and high-dimensional scaling
- Stability Bounds for Stationary φ-mixing and β-mixing Processes
- Rademacher Complexity Bounds for Non-I.I.D. Processes
- Time Series: Theory and Methods
- Forecasting with VARMA Models
- Large Vector Auto Regressions
- Learning Theory and Algorithms for Forecasting Non-stationary Time Series
Cited by
- Multistep Speed Prediction on Traffic Networks: A Graph Convolutional Sequence-to-Sequence Learning Approach with Attention Mechanism
- Time Series Forecasting Using Sequence-to-Sequence Deep Learning Framework
- Classical and Contemporary Approaches to Big Time Series Forecasting
- Streaming Adaptation of Deep Forecasting Models using Adaptive Recurrent Units
- Generalization in fully-connected neural networks for time series forecasting
- Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models
- An LSTM based Encoder-Decoder Model for MultiStep Traffic Flow Prediction
- Estimating Attention Flow in Online Video Networks
- Discrepancy-Based Theory and Algorithms for Forecasting Non-Stationary Time Series
- Neural forecasting: Introduction and literature overview
- Data-Augmentation-Based Cellular Traffic Prediction in Edge-Computing-Enabled Smart City
- Principles and Algorithms for Forecasting Groups of Time Series: Locality and Globality
- Long-range forecasting in feature-evolving data streams
- Multi-step LSTM Prediction Model for Visibility Prediction
- Determinantal Point Processes Implicitly Regularize Semi-parametric Regression Problems
- A Brief Survey of Telerobotic Time Delay Mitigation
- Commentary on “Transparent modelling of influenza incidence”: On big data models for infectious disease forecasting
- Deep Time Series Forecasting With Shape and Temporal Criteria
- Correlated load forecasting in active distribution networks using Spatial‐Temporal Synchronous Graph Convolutional Networks
- Synergetic Learning of Heterogeneous Temporal Sequences for Multi-Horizon Probabilistic Forecasting
Related papers
- Forecasting demand in international markets: The case of correlated time series
- Research on the Time-Data-Separation of the Forecast of Unequal Interval Sequence
- Altering the sequence of time: Narrative technique in O.K. Matsepe's Letšofalela
- Fast Long Sequence Time-Series Forecasting for Edge Service Running State Based on Data Drift and Non-Stationarity
- Backward Sequence Analysis for Single-Armed Cluster Tools
- The remedy for defects of time point sequence theory
- Before" and "Above
- Weight-decreasing reference interval-free continuous DP for retrieval of time-sequence patterns