Pattern recognition and Neural networks

This course provides the opportunity for a student to become familiar with basic concepts that underlie Pattern Recognitions and Neural Networks. The topics cover the fields of Bayesian theory for pattern recognision, decision functions and learning algorithms. Moreover, the basic concepts of network architecture and learning techniques of neural networks are discussed. This purpose is accomplished easily enough if the student follows the course consistently and participate in the solution of the exercises and the development of the tasks assigned.

Objectives

Understanding of the basic principles of Pattern Recognition and Neural Networks

Prerequisites

This information is not available.

Syllabus

Review of linear algebra, linear transformation & probability theory, conditional probability and Bayes rule; Introduction to statistical pattern recognition, feature detection, classification; Bayesian decision theory of pattern recognition; Linear and quadratic discriminant functions; Parametric estimation and supervised learning; Theory of Perceptron; Parzen, K-Near Neighbor (K-NN) classification methods ; Dimensionality reduction, Fisher & entropy techniques; Unsupervised learning, clustering K-means; Neural networks for pattern recognition; Learning