|
Introduction to
neural networks
|
|
Like
nanotechnology, neural networking is the use of technology to
design and manufacture (intelligent) machines, built for
specific purposes, programmed to perform specific tasks. However,
unlike nanomachines, neural networks are designed to work like a
nerve cell system, more similar to the workings of the human or
biological brain in in its physical form. See also
robots and
artificial intelligence.
With today's complex society there is a growing need for
semi-autonomous systems that can do some of the thinking and
controlling for us. The logic of a neural network approximates our
own thinking structures the closest and gives us the opportunity to
endow specific intelligence to designed control systems. See also
cybernetics.
Want to know more? Follow one or several of the free
tutorials on the right to gain knowledge about artificial neural
networks.
|
|
|
|
comp.ai.neural-nets FAQ (7 parts) |
|
Neural Networks Tutorial, Computer Science at the College
of Charleston (US) |
|
Advanced Neural Networks Tutorial, PMSI (F), including
Saxon
4.4 Demos - Prediction, classification, modeling Parabola,
a program to test neural network behavior. |
|
Artificial Neural Networks Technology, Department of
Defense (US) |
|
An Introduction to Neural Networks, Dept. of Electrical &
Computer Engineering, University of Maryland (US) |
|
Crash introduction to Artificial Neural Networks,
University of Massachusetts at Lowell (US) |
|
Neural Networks, Dept. of Computing, Imperial College,
London (UK) |
|
An Introduction to Neural Networks, Centre for Cognitive
and Computational Neuroscience, Department of Computing and
Mathematics, University of Stirling (UK) |
|
|
Neural Network applications |
|
What exactly are neural networks
used for? Artificial neural networks are powerful tools for use
in classification, empirical modeling and pattern recognition,
for example. They are useful in fields as diverse as financing
and investing, business, medical, sports, science and
manufacturing.
They are used to "predict" the rise and fall of stock prices,
race course predictions (horse and dog racing), hospital length
of stay, weather forecasting, earthquake prediction, plastics
and concrete testing, gene recognition.
|
In the field of robotics and artificial intelligence,
artificial neural networks are crucial to the development of the
robotic brain, its logic, its ability to learn, its processing
and analyses of input.
|
|
Neural network software and programming |
|
|
In view of the complexity in designing neural networks it is not
surprising that computers play a major role. No computer without
software and applications made for working with neural networks,
such as design, logic and implementation, are becoming more
plentiful and mainstream. However, this is a growth industry and as
such there always room for writing your own.
|
|
|
|
Joone
- Joone is a free neural network frame, to create, train and
test neural nets. |
|
|
|
|
Neural network hardware |
|
|
On the hardware front of neural network systems great strides
have been made. Mimicking or simulating a neural network can be done
in different ways. The biological approach necessitates the need to
grow and condition or program actual biological nerve cells into
specific behavior.
|
|
|
|
|
|
|
|
|
|
|
|
