Course Identification

1. Dimension reduction, Johnson-Lindenstrauss Lemma and its applications.

2. Compressed sensing and sparse recovery.

3. Inference on large networks: Community detection, stochastic block models and spectral clustering, the planted clique problem.

4. Random matrices, covariance estimation, signal reconstruction in high dimension.

5. Matrix completion.

6. Topics in geometric aspects of machine learning and optimization.

Upon successful completion of the course, the students will become familiar with a variety of mathematical theorems and methods used in the analysis of big data.