Introduction to Model Selection, Decision Theory, Information Theory; Linear Models for Regression and Classification, Neural Networks: Network Training, Jacobian/Hessian Matrices, Regularization, Mixture Density Networks, Bayesian Networks; Computational Learning Theory, Kernel Methods, Sparse Kernel Machines, Graphical Models, Markov Random Fields, Expectation Maximization, Approximate Inference, Factorized Distributions, Expectation Propagation, Hidden Markov Models, Linear Dynamical Systems, Hybrid Model Construction- Boosting, Tree-based models, Conditional Mixture Models, Q-learning and Policy Gradient, PR Applications.