Chapter 11

Recurrent Networks

11.1 Description of Recurrent Neural Network

Based on examples described in earlier chapters, you know that a neural network is explained as a single-layer or multilayer, linear or nonlinear system taught by an algorithm or discovering knowledge on its own. The neural networks discussed to this point are almost free to choose their learning parameters. We say almost because we have not covered recurrent networks yet. These networks include feedback mechanisms that recycle signals from neurons in the input layer to neurons of previously hidden layers (Figure 11.1). Feedback is a relevant and significant innovation. As you will see, a network with feedback has more upgradable possibilities and computing capabilities ...

Get Exploring Neural Networks with C# now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.