Abstract
Artificial neural networks have flourished in recent years in the processing of unstructured data, especially images, text, audio, and speech. Convolutional neural networks (CNNs) work best for such unstructured data. Whenever there is a topology associated with the data, convolutional neural networks do a good job of extracting the important features out of the data. From an architectural perspective, CNNs are inspired by multi-layer Perceptrons. By imposing local connectivity constraints between neurons of adjacent layers, CNN exploits local spatial correlation.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2017 Santanu Pattanayak
About this chapter
Cite this chapter
Pattanayak, S. (2017). Convolutional Neural Networks. In: Pro Deep Learning with TensorFlow. Apress, Berkeley, CA. https://doi.org/10.1007/978-1-4842-3096-1_3
Download citation
DOI: https://doi.org/10.1007/978-1-4842-3096-1_3
Published:
Publisher Name: Apress, Berkeley, CA
Print ISBN: 978-1-4842-3095-4
Online ISBN: 978-1-4842-3096-1
eBook Packages: Professional and Applied ComputingApress Access BooksProfessional and Applied Computing (R0)