ENEE 711 - Neural Networks in Signal Processing

[3]

Fundamentals and characteristics of artificial neural network paradigms and their properties in association, learning, generalization and self-organization; introduction and survey of various neural network models and paradigms, multi-layer perceptron and the radial basis function networks; sum of squares and information-theoretic cost functions; different learning procedures (gradient optimization, conjugate gradients, Newton, etc.); learning and generalization properties; applications in communications and biomedical signal processing; and comparisons with linear adaptive signal processing theory and techniques.
Prerequisite: Prerequisite: ENEE 620 or consent of instructor.