×

TPA-LSTM

swMATH ID: 34694
Software Authors: Shih, Shun-Yao; Sun, Fan-Keng; Lee, Hung-yi
Description: Temporal pattern attention for multivariate time series forecasting. Forecasting of multivariate time series data, for instance the prediction of electricity consumption, solar power production, and polyphonic piano pieces, has numerous valuable applications. However, complex and non-linear interdependencies between time steps and series complicate this task. To obtain accurate prediction, it is crucial to model long-term dependency in time series data, which can be achieved by recurrent neural networks (RNNs) with an attention mechanism. The typical attention mechanism reviews the information at each previous time step and selects relevant information to help generate the outputs; however, it fails to capture temporal patterns across multiple time steps. In this paper, we propose using a set of filters to extract time-invariant temporal patterns, similar to transforming time series data into its “frequency domain”. Then we propose a novel attention mechanism to select relevant time series, and use its frequency domain information for multivariate forecasting. We apply the proposed model on several real-world tasks and achieve state-of-the-art performance in almost all of cases. Our source code is available at url{https://github.com/gantheory/TPA-LSTM}.
Homepage: https://arxiv.org/abs/1809.04206
Source Code:  https://github.com/gantheory/TPA-LSTM
Keywords: multivariate time series; attention mechanism; recurrent neural network; convolutional neural network; polyphonic music generation
Related Software: DeepAR; PyTorchTS; Faster R-CNN; Adam; GitHub; AlexNet; MuseGAN; ImageNet
Cited in: 3 Documents

Citations by Year