Course Number:

Credit Hours:

4

Course Title:

Pattern Recognition

Course Description:

Discusses introductory concepts, statistical classification problem, and the Bayes classifier. Covers parametric estimation and supervised learning, ML and Bayes approaches, and Bayes learning. Topics include nonparametric techniques, Parzen windows, nearest neighbor rules, convergence properties, and error bounds. Examines linear discriminant functions, linear separability, perceptrons and their training, and relaxation techniques. Discusses unsupervised learning and clustering, unsupervised Bayes learning, ML estimates, k-means algorithm, and learning vector quantization. Introduces neural network structures, feed-forward nets, ADALINE, Widrow-Hopf approach, the back propagation training algorithm, Kolmogorov’s theorem, and capacity of feed-forward nets. Focuses on Hopfield model and learning, associative memory, bidirectional associative memory, stable states and convergence, and capacity of the Hopfield model. Also covers unsupervised learning, adaptive resonance theory, and self-organizing feature maps.