Natural evolution is a population-based optimization process. Computer simulation of natural evolution results in stochastic optimization techniques that surpass traditional optimization methods. This course will provide a background in the inspiration, history, and application of evolutionary computation methods with particular emphasis on problems within signal processing. System identification, pattern recognition, automatic control, and gaming demonstrate useful hybridization between evolutionary computation and fuzzy and neural systems.

This course provides attendees with an introduction to the fundamentals of neural networks and evolutionary computation and the manner in which they can be applied to applications in homeland security. The course concentrates on applying computational intelligence methods to pattern discovery and recognition in security problems with examples in areas including risk assessment, physical security, and personal identification. You will learn how to apply computational intelligence methods to problems and how to choose appropriate design parameters.

This course is an introduction to the central concepts that underlie evolutionary and neural computation and describes how to apply these methods to typical problems in signal processing, including time series prediction, classification, clustering, and control. Fundamental issues involving data representation, search operations, selection, and other facets of problem solving are addressed.