LSTM Fully Convolutional Networks for Time Series Classification

TLDR

Fully convolutional neural networks (FCN) have achieved state‑of‑the‑art performance on time‑series classification tasks. The authors propose augmenting FCNs with LSTM sub‑modules, adding an attention mechanism, and fine‑tuning to improve time‑series classification. The method combines FCNs with LSTM sub‑modules, incorporates an attention mechanism, applies fine‑tuning, and performs comparative analysis against other techniques. The LSTM‑FCN and its attention‑augmented variant outperform existing methods, improve performance with minimal preprocessing, and enable visualization of the LSTM decision process.

Abstract

Fully convolutional neural networks (FCN) have been shown to achieve state-of-the-art performance on the task of classifying time series sequences. We propose the augmentation of fully convolutional networks with long short term memory recurrent neural network (LSTM RNN) sub-modules for time series classification. Our proposed models significantly enhance the performance of fully convolutional networks with a nominal increase in model size and require minimal preprocessing of the dataset. The proposed Long Short Term Memory Fully Convolutional Network (LSTM-FCN) achieves state-of-the-art performance compared to others. We also explore the usage of attention mechanism to improve time series classification with the Attention Long Short Term Memory Fully Convolutional Network (ALSTM-FCN). Utilization of the attention mechanism allows one to visualize the decision process of the LSTM cell. Furthermore, we propose fine-tuning as a method to enhance the performance of trained models. An overall analysis of the performance of our model is provided and compared to other techniques.