Artifical Neural Networks

Doing research for a school assignment just now, I found an article on Artificial Neural Networks.

An Artificial Neural Network is an attempt to simulate the human brain, and, when used with a process called Back Propagation is much better at identifying patterns than a computer built upon rules of logic. Back Propagation is a method in which the computer is given a large set of things to look for in an image, and then told to complete that command. At first, the results are random, because the sets are random, but when the results are processed, it can be determined which sets of points are significant to the task of the Neural Network. In this way, it is easy to correct the computer, so that in the next cycle, it focuses more on the important aspects of the images. Eventually, a clear definition of what to search for can be developed, and much more easily than it would be to do the same with a logical computer, creating strict laws for it to search for. Essentially, an Artificial Neural Network is capable of training itself.

I find this fascinating, and it seems like a much more feasible way to create an artificial intelligence than most. I highly encourage you to read the original article, which you can find here.