Insights into LSTM Fully Convolutional Networks for Time Series Classification

Details

• Personal Author:
  Darabi H ; Karim F ; Majumdar S
• Description:
  Long short-term memory fully convolutional neural networks (LSTM-FCNs) and Attention LSTM-FCN (ALSTM-FCN) have shown to achieve the state-of-the-art performance on the task of classifying time series signals on the old University of California-Riverside (UCR) time series repository. However, there has been no study on why LSTM-FCN and ALSTM-FCN perform well. In this paper, we perform a series of ablation tests (3627 experiments) on the LSTM-FCN and ALSTM-FCN to provide a better understanding of the model and each of its sub-modules. The results from the ablation tests on the ALSTM-FCN and LSTM-FCN show that the LSTM and the FCN blocks perform better when applied in a conjoined manner. Two z-normalizing techniques, z-normalizing each sample independently and z-normalizing the whole dataset, are compared using a Wilcoxson signed-rank test to show a statistical difference in performance. In addition, we provide an understanding of the impact dimension shuffle that has on LSTM-FCN by comparing its performance with LSTM-FCN when no dimension shuffle is applied. Finally, we demonstrate the performance of the LSTM-FCN when the LSTM block is replaced by a gated recurrent unit (GRU), basic neural network (RNN), and dense block. [Description provided by NIOSH]
• Subjects:
  Occupational Health Safety Software
• Keywords:
  Author Keywords: Convolutional Neural Network Computer Applications Computer Software Industrial Engineering Long Short Term Memory Recurrent Neural Network Neural Networks Time Series Classification
• ISSN:
  2169-3536
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Journal Article
• Place as Subject:
  Illinois ; OSHA Region 5
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Volume:
  7
• NIOSHTIC Number:
  nn:20063644
• Citation:
  IEEE Access 2019 May; 7:67718-67725
• Contact Point Address:
  Houshang Darabi, Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL 60607, USA
• Email:
  hdarabi@uic.edu
• Federal Fiscal Year:
  2019
• Performing Organization:
  University of Illinois at Chicago
• Peer Reviewed:
  True
• Start Date:
  20050701
• Source Full Name:
  IEEE Access
• End Date:
  20290630
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:d0e95241f4d2ad8769e33c48dc87e9dd88bdddaeb011f70e8f7e02cc7312ee23f6cb92f3e84688b6c385b16e0e36d903649edeb0fdbce6f2c60281af8c454462
• Download URL:
  https://stacks.cdc.gov/view/cdc/222212/cdc_222212_DS1.pdf
• File Type:
  [PDF - 5.92 MB ]