This course discusses some of the issues (such as dimensionality reduction, supervised/unsupervised learning, etc) of machine learning from the perspective of optimization.  Broadly, the course aims at covering the following topics:

• Convex Optimization: A quick review of convex optimization problems, Descent methods, KKT conditions,  Duality

• Dimensionality reduction: Linear and nonlinear dimensionality reduction techniques (such as principal component analysis, linear discriminant analysis, method of random projections, multidimensional scaling, manifold learning)

• Classification: Linear and nonlinear support vector machines, Kernels, Neural networks

• Clustering: K-means algorithm, its variants and cluster evaluation procedures. 

Pre-requisite:  Linear algebra