Precise asymptotics in high-dimensional sparse regression and compressed sensing.

David Donoho
Department of Statistics
Stanford University


Wednesday, May 11, 2011
4:30 - 5:30 PM
Packard 101

Abstract:

I will describe recent work giving precise asymptotic results on mean squared error and other characteristics, of a range of estimators in a range of high-dimensional problems from sparse regression and compressed sensing; these include results for LASSO, group LASSO, and nonconvex sparsity penalty methods. A key applicaton of such precise formulas is their use in deriving precise optimality results which were not known previously, and to our knowledge not available by other methods.

This is joint work over several papers with several co-authors, including Andrea Montanari, Iain Johnstone, and Arian Maleki.

---

Operations Research Colloquia: http://or.stanford.edu/oras_seminars.html