Compressed Sensing and High-Dimensional Linear Regression
NCMIS Distinguished Lectures

Presented at the Chinese Academy of Sciences, Beijing, June 2014


Abstract: These lectures will focus on compressed sensing and high dimensional linear regression. These and other related problems have attracted much recent interest in a range of fields including statistics, machine learning and electrical engineering. In the high dimensional setting where the dimension p can be much larger than the sample size n, classical methods and results based on fixed p and large n are no longer applicable. We will analyze in detail the constrained and penalized l1 minimization methods for compressed sensing/high-dimensional regression and give a unified and elementary analysis on sparse signal recovery in three settings: noiseless, bounded noise and Gaussian noise. Johnson-Lindenstrauss Lemma and construction of compressed sensing matrices will also be discussed. In addition, we will discuss at the end some recent results on optimal and adaptive estimation of high-dimensional covariance/precision matrices under different settings.


References:


Back to Tony Cai's Homepage