Neural networks are a powerful technique to solve many real world problems. They have the ability to learn from experience in order to improve their performance and to adapt themselves to changes in the environment. In addition to that they are able to deal with incomplete information or noisy data and can be very effective, especially in situations where it is not possible to specify rules or steps that lead to the solution of a problem.
UReason's Neural Network capabilities allow you to construct custom Neural Networks from a preconfigured set of neural components.
Neural Networks are typically used to calculate unmeasured values i.e. act a soft sensor; and/or forecast values.The types of Neural Networks supported by UReason are:
- Supervised Neural Networks
- Modular Neural Networks;
- Temporal Feed Forward Neural Networks; and
- Unsupervised Neural Networks.
The learning algorithms used for Supervised Neural Networks are:
- The basic online back propagation algorithm
- The batch back propagation algorithm
- The resilient back propagation algorithm
The learning algorithms used for Unsupervised Neural Networks are:
- The PCA algorithm (Principle Component Analysis) in the Sanger Synapse.
- The Self Organising Maps algorithm in the Kohonen Synapse.
A Neural network can be considered as a black box that is able to predict an output pattern when it recognises a given input pattern. Once trained, the neural network is able to recognise similarities when presented with a new input pattern, resulting in a predicted output pattern.
The most important aspect of a Neural Network is the learning process. Learning can be accomplished by supervised or unsupervised training.
In supervised training, both the inputs and the outputs are provided. The network then processes the inputs and compares its resulting outputs against the desired outputs. Errors are then calculated, causing the system to adjust the weights which control the network. This process occurs over and over as the weights are continually tweaked.
In unsupervised training, the network is provided with inputs but not with desired outputs. The system itself must then decide what features it will use to group the input data. This is often referred to as self-organisation or adaption.