Continuously Recombinant Genetic Neural Fiber Networks
James A. Crowder · AIAA Infotech@Aerospace 2010 · 2010
The underlying issues and challenges posed by the introduction of Artificial Intelligence into system designs are not new. Information processing and dissemination systems are an expensive infrastructure to operate and moreoften-than-not these systems fail to provide analysts with tangible and useful situational information, typically overwhelming information analysts with system messages and other low-level data. Real-time human decision making processes must be supported by information derived from the fusion process and must operate in a uniform and cooperative model, fusing data into information and knowledge, so information analysts can make informed decisions. What is described here is a modular architecture based on a mixture of neural structures that add flexibility and diversity to the overall system capabilities. Presented here is the object architecture for a flexible, continually adaptable neural processing system capable of dynamically adding and pruning basic building blocks of the neural system as the real-time requirements of the system change. This defines the object architecture for the Evolving, Life-like Yielding Symbiotic Environment (ELYSE) system (Crowder, 2001_002). This modular architecture is based on a “mixture of experts” methodology. The difference here is that in our architecture, an expert is defined as a particular fuzzy, genetic perceptron object which has been created for a particular algorithm, and thus is an expert at processing a particular type of data in a particular manner. The algorithm for which the perceptron is generated may be predetermined or may have been evolved by the neural system itself and provides a continuously evolutionary system architecture based on genetic learning within the recombinant neural structure.