Abstract

Scientists and Engineers constantly face the challenge of trying to find solutions to problems which are not well understood, complex, swamped in noisy data or a combination of negative factors. Research by its very nature involves working with systems for which we don't yet have an accurate (or any!) model, systems which are generally multivariable, high order, and nonlinear. If we now throw noisy data measurements into the mix, then even relatively simple problems become intractable by 'classical methods'.
In this seminar, Professor Stewart examines problem solving methodologies, with particular emphasis on Biologically Inspired Heuristics and Meta-Heuristics, with particular emphasis on data analysis and modeling with real-world applications.