Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks

Multivariate time series forecasting is an important machine learning problem across many domains, including predictions of solar plant energy output, electricity consumption, and traffic jam situation. Temporal data arise in these real-world applications often involves a mixture of long-term and short-term patterns, for which traditional approaches such as Autoregressive models and Gaussian Process may fail. In this paper, we proposed a novel deep learning framework, namely Long- and Short-term Time-series network (LSTNet), to address this open challenge. LSTNet uses the Convolution Neural Network (CNN) and the Recurrent Neural Network (RNN) to extract short-term local dependency patterns among variables and to discover long-term patterns for time series trends. Furthermore, we leverage traditional autoregressive model to tackle the scale insensitive problem of the neural network model. In our evaluation on real-world data with complex mixtures of repetitive patterns, LSTNet achieved significant performance improvements over that of several state-of-the-art baseline methods. All the data and experiment codes are available online.

[1]  Jürgen Schmidhuber,et al.  Highway Networks , 2015, ArXiv.

[2]  Deng Cai,et al.  What to Do Next: Modeling User Behaviors by Time-LSTM , 2017, IJCAI.

[3]  Les E. Atlas,et al.  Recurrent Networks and NARMA Modeling , 1991, NIPS.

[4]  A. Banerjee,et al.  Estimating Structured Vector Autoregressive Models , 2016, ICML.

[5]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[6]  S Roberts,et al.  Gaussian processes for time-series modelling , 2013, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[7]  Ugur Demiryurek,et al.  Deep Learning: A Generic Approach for Extreme Condition Traffic Forecasting , 2017, SDM.

[8]  Kyoung-jae Kim,et al.  Financial time series forecasting using support vector machines , 2003, Neurocomputing.

[9]  Shih-Chii Liu,et al.  Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences , 2016, NIPS.

[10]  Xiaoli Li,et al.  Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition , 2015, IJCAI.

[11]  Fang Han,et al.  Robust Estimation of Transition Matrices in High Dimensional Heavy-tailed Vector Autoregressive Processes , 2015, ICML.

[12]  Roger Frigola,et al.  Bayesian Time Series Learning with Gaussian Processes , 2015 .

[13]  Jonathan D. Cryer,et al.  Time Series Analysis , 1986 .

[14]  Carl E. Rasmussen,et al.  Bayesian Inference and Learning in Gaussian Process State-Space Models with Particle MCMC , 2013, NIPS.

[15]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[16]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1971 .

[17]  Francis Eng Hock Tay,et al.  Support vector machine with adaptive parameters in financial time series forecasting , 2003, IEEE Trans. Neural Networks.

[18]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[19]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .

[20]  Yoshua Bengio,et al.  Convolutional networks for images, speech, and time series , 1998 .

[21]  Yoshua Bengio,et al.  Neural Machine Translation by Jointly Learning to Align and Translate , 2014, ICLR.

[22]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups , 2012, IEEE Signal Processing Magazine.

[23]  Alexander J. Smola,et al.  Support Vector Method for Function Approximation, Regression Estimation and Signal Processing , 1996, NIPS.

[24]  Thomas Plötz,et al.  Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables , 2016, IJCAI.

[25]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[26]  Guoqiang Peter Zhang,et al.  Time series forecasting using a hybrid ARIMA and neural network model , 2003, Neurocomputing.

[27]  Takayuki Osogami,et al.  Nonlinear Dynamic Boltzmann Machines for Time-Series Prediction , 2017, AAAI.

[28]  Inderjit S. Dhillon,et al.  Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction , 2016, NIPS.

[29]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[30]  E. McKenzie General exponential smoothing and the equivalent arma process , 1984 .

[31]  Ashu Jain,et al.  Hybrid neural network models for hydrologic time series forecasting , 2007, Appl. Soft Comput..

[32]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[33]  Jiahan Li,et al.  Forecasting Macroeconomic Time Series: LASSO-Based Approaches and Their Forecast Combinations with Dynamic Factor Models , 2014 .

[34]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[35]  G. Box,et al.  Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models , 1970 .

[36]  Gregory D. Hager,et al.  Temporal Convolutional Networks: A Unified Approach to Action Segmentation , 2016, ECCV Workshops.

[37]  Yoshua Bengio,et al.  Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling , 2014, ArXiv.

[38]  Charles Elkan,et al.  Learning to Diagnose with LSTM Recurrent Neural Networks , 2015, ICLR.

[39]  Yan Liu,et al.  Recurrent Neural Networks for Multivariate Time Series with Missing Values , 2016, Scientific Reports.