Course Description: This course introduces fundamental concepts, theories, and algorithms for pattern recognition and machine learning, which are used in computer vision, speech recognition, data mining, statistics, information retrieval, and bioinformatics. Topics include but are not limited to, various parametric and nonparametric methods for supervised and unsupervised learning problems.
Prerequisites: Basic calculus, probability & random processes, linear algebra. A basic understanding of signal processing is suggested.
Course contents:
Introduction to PRML: General Notions, Parameter estimation, overfitting, model selection, the curse of dimensionality, bias-variance tradeoff;
Supervised Learning (Regression & Classification): Density estimation, Bayes decision theory, generative vs. discriminative models, Linear Methods: linear & logistic regression, generalized linear models, linear discriminant functions for classification, support vector machines, etc., Nonlinear methods: kernel methods, nearest neighbor, neural networks, etc.,
Unsupervised Learning (Clustering & Density Estimations): K-means clustering, vector quantization, Gaussian mixture models, autoencoders, dimensionality reduction (linear & nonlinear)
Handling Sequential Data: Hidden Markov models, and Linear Dynamical systems.
References:
CM Bishop, Pattern Recognition and Machine Learning, Springer
R O Duda, P E Hart, Pattern Classification, Wiley
Kevin P Murphy, Machine Learning: A probabilistic perspective, MIT Press