Page 228 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

CHAPTER

Deep Learning

Approaches to 11

Electrophysiological

Multivariate Time-Series

Analysis*

Francesco Carlo Morabito, Maurizio Campolo, Cosimo Ieracitano, Nadia Mammone
NeuroLab, DICEAM, University Mediterranea of Reggio Calabria, Reggio Calabria, Italy

CHAPTER OUTLINE
1. Introduction .......................................................................................................220
2. The Neural Network Approach.............................................................................220
3. Deep Architectures and Learning.........................................................................222
3.1 Deep Belief Networks .......................................................................... 223
3.2 Stacked Autoencoders ......................................................................... 224
3.3 Convolutional Neural Networks ............................................................. 224
4. Electrophysiological Time-Series ........................................................................226
4.1 Multichannel Neurophysiological Measurements of the
Activity of the Brain............................................................................. 226
4.2 Electroencephalography (EEG) ............................................................. 226
4.3 High-Density Electroencephalography ................................................... 228
4.4 Magnetoencephalography..................................................................... 232
5. Deep Learning Models for EEG Signal Processing.................................................232
5.1 Stacked Autoencoders ......................................................................... 232
5.2 Summary of the Proposed Method for EEG Classification........................ 235
5.3 Deep Convolutional Neural Networks..................................................... 236
5.4 Other DL Approaches........................................................................... 236
6. Future Directions of Research .............................................................................236
6.1 DL Interpretability............................................................................... 238
6.2 Advanced Learning Approaches in DL ................................................... 238
6.3 Robustness of DL Networks.................................................................. 239
7. Conclusions.......................................................................................................239
References .............................................................................................................240
Further Reading ......................................................................................................243

* To my loved daughter, Valeria
219
Artificial Intelligence in the Age of Neural Networks and Brain Computing. https://doi.org/10.1016/B978-0-12-815480-9.00011-6
Copyright © 2019 Elsevier Inc. All rights reserved.