Neural Networks - Potential application in the nuclear industry: Preliminary comments and reflections

Neural Networks (or Parallel Distributed Processing, Connectionist Models, Layered Self-Adaptive Systems, Self-Organizing Systems) are an emerging technology which is perceived to have potential for solving complex computation problems which cannot be solved by standard computational methods. A neural network is a system composed of many simple processing elements operating in parallel whose function is determined by network structure, connection strengths and the processing performed at computing elements. Neural networks' architectures are inspired by the architecture of biological nervous systems which use a large number of simple processing elements, the neurons, operating in parallel to obtain high computation rates. Neural networks can learn and can be taught. They learn from examples, adjusting internal parameters to match the examples. Neural networks can be considered as a different approach to computation. In the future, neural networks' modules may possibly be embedded in expert systems and conventional applications. Thus, the three information processing technologies, procedural languages, expert system and neural networks, running on the same hardware, can be used either separately, to solve specific types of problems, or in various combinations where each brings its unique capabilities to the overall application