Applications & Learning

Getting the networks to do something is a matter of giving them inputs, and letting the information travel through the topology by simulating each artificial neuron.

Applications

Once inserted into more complex systems, the ouputs can then be used to solve a problem:

• Classification - Detecting patterns in the input, and determining what class it belongs to.
  Sorting chocolates based on visual appearance
• Prediction - Given the input, decide what's going to happen next!
  Trying to predict the stock exchange!
• Regression - Provided with sensor inputs, the neural network can combine them in a smooth output.
  Controlling a robotic arm with motors based on detectors

Learning

How does the neural network know how to perform? Basically it needs to learn, which explains why this is a field of machine learning.

• Training - Based on a set of examples, it can try to understand the main idea.
• Self-Organisation - Using complex models of cooperation and competition between neurons, some networks can organise themselves over time!
• Evolution - The concept of the survival of the fittest applied to obtaning a good solution.

Remember you can visit the Message Store to discuss this tutorial.
Warning: mysql_connect() [function.mysql-connect]: Access denied for user 'alexjc_r'@'209.68.2.84' (using password: YES) in /usr/home/alexjc/public_www/Include/Utils.php on line 5
Access denied for user 'alexjc_r'@'209.68.2.84' (using password: YES)