Biologically Rationalized Computing Techniques For Image Processing Applications by Jude Hemanth & Valentina Emilia Balas

Author:Jude Hemanth & Valentina Emilia Balas
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

2. Deep networks for supervised learning (labeled data)

3. Hybrid deep networks (combination of 1 and 2).

4.1 Unsupervised Learning

Unsupervised learning is advantageous in cases where there is a huge amount of unlabeled data. By applying unsupervised deep learning techniques to such data, features that are better compared to handcrafted features can be learned.

Autoencoder. An autoencoder is used in unsupervised deep learning techniques. An autoencoder consists of an encoder followed by a decoder as shown in Fig. 8. The encoder part transforms the input x to z using h e(.). The decoder part tries to get back the original input x from z. The decoder outputs which is the approximate reconstruction of the original input x.

Fig. 8Autoencoder

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.