Course content

• Supervised learning: decision trees, nearest neighbor classifiers, generative classifiers like naive Bayes, linear discriminant analysis, loss regularization framework for classification, Support vector Machines Regression methods: least-square regression, kernel regression, regression trees Unsupervised learning: k-means, hierarchical, EM, non-negative matrix factorization, rate distortion theory.

References

• Hastie, Tibshirani, Friedman The elements of Statistical Learning Springer Verlag
• Pattern recognition and machine learning by Christopher Bishop.
• Selected papers.