To explain the concepts of machine learning and its applications.
To discuss supervised learning
To make rational decisions to minimize expected risk using Bayesian decision theory
To estimate the probabilities from a given training set using parametric and non parametric methods
To discuss learning algorithms used for linear discrimination
To analyze multilayer perceptron used for classification and regression
Course Contents
UNIT I
Introduction: Probability theory, what is machine learning, example machine learning applications
Supervised Learning: Learning a class from examples, VC dimension, PAC learning, Noise, Learning multiple classes, Regression, Model selection and generalization
UNIT II
Bayesian Learning: Classification, losses and risks, utility theory MLE, Evaluating an estimator, Bayes estimator, parametric classification
Discriminant functions: Introduction, Discriminant functions, Least squares classification, Fisher’s linear discriminant, fixed basis functions, logistic regression
Non-parametric methods: Nearest Neighbor Classifier, Non-parametric Density Estimation
UNIT IV
Maximum margin classifiers: SVM, Introduction to kernel methods, Overlapping class distributions, Relation to logistic regression, Multiclass SVMs, SVMs for regression Mixture models and EM: K – means clustering, Mixture of Gaussians, Hierarchical Clustering, Choosing the Number of Clusters
UNIT V
Dimensionality reduction: Combining Model Regression with sampling, Bayes classifier, Perceptron algorithm and clustering algorithms.
Books
Bishop, C. M. (2006). Pattern recognition and machine learning. springer.
Ethem Alpaydin. (2010) Introduction to Machine Learning, Second Edition, PHI Learning Pvt. Ltd