De-mixing problems in spectroscopic imaging often require finding sparse non-negative linear combinations of library functions to match observed data. Due to misalignment and uncertainty in data measurement, the known library functions may not represent the data as well as their proper deformations. To improve data adaptivity, we expand the library to one with a group structure and impose a structured sparsity constraint so that the coefficients for each group should be sparse or even 1-sparse. Since the expanded library is a highly coherent (redundant) dictionary, it is difficult to obtain good solutions using convex methods such as non-negative least squares (NNLS) or L1 norm minimization. We study efficient non-convex penalties such as the ratio/difference of L1 and L2 norms, as sparsity penalties to be added to the objective in NNLS-type models. We show an exact recovery theory of the sparsest solution by minimizing the ratio/difference norms under a uniformity condition. For solving the related unconstrained non-convex models, we develop a scaled gradient projection algorithm that requires solving a sequence of strongly convex quadratic programs.

I will use concrete examples to argue why mathematics is even more important and powerful when computers become more and more powerful and to show what computational mathematics is about.

Which natural numbers occur as the area of a right triangle with three rational sides? This is a very old question and is still unsolved, although partial answers are known (for example, five is the smallest such natural number). In this talk we will discuss this problem and recent progress that has come about through its connections with elliptic curves and other important open questions in number theory.

By now there is a long list of questions in analysis and algebra which are known to be undecided in the standard set theory (Zermelo-Fraenkel). In particular, no standard methods accepted and used by mathematicians can provide a proof deciding such questions. Yet, a definitive answer is often desirable. I will discuss some axioms that settle most of these open questions, provide useful extensions of standard set theory, and are intersting on their own. These axioms rely on the existence of sets that are significantly "larger" than any sets mainstream mathematics works with. 

In this talk, I will present some rigidity problems and theorems from analysis, partial differential equations and differential geometry. For examples, the uniqueness theorem of holomorphic functions upper rigidity of harmonic mapping. In particular, I will present some rigidity theorem for proper holomorphic mapping and smooth solutions of some degenerate elliptic partial differential equations.