Pattern Recognition and Computational Intelligence Techniques Using Matlab

The primary objective of this book is to make the mathematical foundations of pattern recognition and computational intelligence easily accessible. It provides comprehensive coverage of various regression methodologies, including Bayesian techniques, Maximum Likelihood, Least Squares, regularization methods, kernel smoothing, and Gaussian processes, alongside Evidence Approximation, Support Vector Regression (SVR), and Relevance Vector Regression (RVR). The text further explores essential dimensionality reduction techniques such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Kernel LDA, and Independent Component Analysis (ICA). Readers will find detailed treatments of probabilistic generative models like Gaussian Mixture Models (GMM) and Hidden Markov Models (HMM), the generalized Expectation-Maximization (EM) algorithm, and probabilistic discriminative models including Support Vector Machines (SVM) and multi-class logistic regression. Additionally, the book covers vital computational intelligence metaheuristics—specifically Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO)—and introduces various deep learning algorithms in addition to outlining statistical testing relevant to pattern recognition. Building upon the core topics of the first edition, this second edition introduces entirely new material on sampling techniques and graphical models. The sampling section covers rejection sampling, Gibbs sampling, and importance sampling. The newly added section on graphical models discusses computing marginal probabilities via the sum-product algorithm and details methods for identifying optimal random variable values that maximize joint probability. Finally, the chapters on linear regression and probabilistic generative models have been thoroughly revised and updated for this edition.