A Memory-Network Based Solution for Multivariate Time-Series Forecasting

Abstract

Multivariate time series forecasting is extensively studied throughout the years with ubiquitous applications in areas such as finance, traffic, environment, etc. Still, concerns have been raised on traditional methods for incapable of modeling complex patterns or dependencies lying in real word data. To address such concerns, various deep learning models, mainly Recurrent Neural Network (RNN) based methods, are proposed. Nevertheless, capturing extremely long-term patterns while effectively incorporating information from other variables remains a challenge for time-series forecasting. Furthermore, lack-of-explainability remains one serious drawback for deep neural network models. Inspired by Memory Network proposed for solving the question-answering task, we propose a deep learning based model named Memory Time-series network (MTNet) for time series forecasting. MTNet consists of a large memory component, three separate encoders, and an autoregressive component to train jointly. Additionally, the attention mechanism designed enable MTNet to be highly interpretable. We can easily tell which part of the historic data is referenced the most.

arXiv comment8 pages, 4 figures, submitted to AAAI 2019. arXiv admin note: text overlap with arXiv:1703.07015 by other authors

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction
• Dynamic Covariance Models for Multivariate Financial Time Series
• Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks
• Multi-variable LSTM neural network for autoregressive exogenous model

Cited by in corpus (7)

• Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
• A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting
• LAVARNET: Neural Network Modeling of Causal Variable Relationships for Multivariate Time Series Forecasting
• Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values
• A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting
• From Known to Unknown: Knowledge-guided Transformer for Time-Series Sales Forecasting in Alibaba
• Selecting Datasets for Evaluating an Enhanced Deep Learning Framework