One of the more exiting fields in neurotechnology, is that of neural networks. Even as early as the middle of the 20th century, observers such as Turing (who was responsible for the famous Turing Test) noted that the brain functioned by means of components whose connections were responsible for complex processing.

Current methods of solving problems are procedural and object oriented. Meaning that we try and break a problem down into component parts and assign various functions to different objects. For example, within a CPU, we have the ALU for Arithmetic functions. The neurons the brain however, seem to perfom work in parallel in a sort of distributed manner.

Complex phenomena like emotions and awareness are the product of dynamic connections between neurons that are adaptive in nature. Even memories are said to be represented by connection patterns.

The field of neural networks seeks to learn from the brain in order to represent these neurons in an artificial manner. Though mathematical models currently leave a lot to be desired, we are able to replicate a few characteristics of the brain.

Artificial neural networks (ANNs) are currently used in the real world to help with regression problems (finding a function that fits the given data) as well as pattern and sequence recognition. These can be adapted for uses as wide as creating efficient spam filters or creating a medical diagnosis.

This blog covers the latest news and technologies in neural networks and provides the reader with an easy understanding of the technological implications.