Machine Learning. 3 credits

Catalog description

This course covers the theory and principles underlying different machine learning paradigms. The emphasis is on statistical theory and methodology. Topics include: Model selection and generalization; Overfitting and under fitting; Bayesian theory and Decision theory; Maximum Likelihood estimation, MAP; Regularization; Bias-variance tradeoff; Curse of dimensionality; Dimensionality reduction; Linear Models for classification; Probabilistic Generative Models; Probabilistic Discriminative Models; Neural Networks (Backpropagation); Deep Learning (CNNs); Kernel methods; Support Vector Machines; Ensemble Methods; Unsupervised Learning (Clustering, EM, Mixture Modeling); Reinforcement learning. Offered by Computer Science . May not be repeated for credit.