C# Elman Neural Network

The Elman neural network is a simple recurrent neural network (SRN) developed by Jeffrey L. Elman in 1990. This network type consists of an input layer, a hidden layer, and an ouput layer. In this way it resembles a three layer feedforward neural network. However, it also has a context layer. This context layer is fed, without weighting, the output from the hidden layer. The Elman network then remembers these value and outputs them on the next run of the neural network. These values are then sent, using a trainable weighted connection, back into the hidden layer. Elman neural networks are very useful for predicting sequences, since they have a limited short-term memory.

Elman Nueral networks are described in "Finding Structure in Time", Cognitive Science 14, 179-211 (1990), by Jeffrey L. Elman.

The Elman Neural Network modeled in the Encog Workbench looks like this: