Pattern Recognition

Introduction: Overview of pattern recognition, review of matrix algebra, probability distributions and probability; Bayesian Decision Theory: Classifiers, discriminant functions and decision surfaces, normal density and discriminant functions for normal density, error bounds for normal densities; Bayesian Parameter Estimation: Univariate and Multivariate cases for the Gaussian distribution, bias, class-conditional densities; Maximum Likelihood Estimation: The general principle, Unknown parameter cases for Gaussian distribution; Problems of dimensionality: Accuracy and training sample size, computational complexity, overfitting; Component Analysis: Principle Component Analysis, Fisher Linear discriminant; Non-Parametric Techniques: Parzen windows, K-Nearest neighbour rule and estimation.