Events
Daniel Bernstein, NC State, Unimodular hierarchical models
The Zariski closure of a discrete log-linear statistical model is a toric variety. In the last twenty years, this fact has lead not only to the development of useful algorithms for working with data, but developments in algebraic geometry and combinatorics as well. I will discuss a particular subset of the discrete log-linear models known…
What is IBL? (part 2)
SAS 1102We'll start with a panel of graduate instructors currently experimenting with IBL methods in their courses. We will then discuss and brainstorm ways to effectively get students to share their answers and ideas during lecture. For those that missed the first session, the following article gives an introduction to IBL: http://maamathedmatters.blogspot.com/2013/05/what-heck-is-ibl.html
Let’s Talk IT: Moodle
SAS 4201Chris from DELTA will be leading an interactive seminar on what Moodle is, a jump start, and answering any questions you have for the platform and it's capabilities. You'll have the opportunity to go hands on with them as they walk you through the platform. Regardless of your experience with Moodle, I encourage you to…
Gabor Pataki, UNC-Chapel Hill, Bad semidefinite programs, linear algebra, and short proofs
SAS 4201Semidefinite programs (SDPs) -- optimization problems with linear constraints, linear objective, and semidefinite matrix variables -- are some of the most useful, versatile, and pervasive optimization problems to emerge in the last 30 years. They find applications in combinatorial optimization, machine learning, and statistics, to name just a few areas. Unfortunately, SDPs often behave pathologically: the optimal values of the primal…
Weiwei Hu, Oklahoma State University, An Approximating Control Design for Optimal Mixing by Stokes Flows
SAS 4201We consider an approximating control design for optimal mixing of a non-dissipative scalar eld in unsteady Stokes ows. The objective of our approach is to achieve optimal mixing at a given nal time, via an active control of the ow velocity through boundary inputs. Due to the zero diusivity of the scalar eld, establishing the well-posedness of its…
Zack Morrow, NC State, SIAM Student Chapter Tutorial Series, Sparse methods for high-dimensional problems
SAS 2235In full-tensor extensions of 1D interpolation or quadrature rules, the number of nodes will grow exponentially in the dimension—commonly called the “curse of dimensionality.” In this talk, we present an overview of sparse grids, in which the number of nodes grows only polynomially in the dimension. First, we will do an overview of 1D interpolation for…
Jane Coons, NC State, Geometry and Combinatorics of the Cavander-Farris-Neyman Model with a Molecular Clock
We prove results about the polytope associated to the toric ideal of invariants of the Cavender-Farris-Neyman model with a molecular clock on a rooted phylogenetic tree. For instance, the number of vertices of this polytope is a Fibonacci number and the facets of the polytope can be described using the combinatorial structure of the underlying…
Tea and Cookies
SAS 4104Michael Overton, NYU, Nonsmooth, Nonconvex Optimization: Algorithms and Examples
SAS 1102In many applications one wishes to minimize an objective function that is not convex and is not differentiable at its minimizers. We discuss two algorithms for minimization of nonsmooth, nonconvex functions. Gradient Sampling is a simple method that, although computationally intensive, has a nice convergence theory. The method is robust and the convergence theory has…
Radmila Sazdanović to speak at Coffee and Viz Series
James B. Hunt Jr. Library at Centennial CampusRadmila Sazdanović will speak at the Coffee and Viz series in the Teaching and Visualization Lab, James B. Hunt Jr. Library on Friday March 16, 9:30-10:30 am about Visual Mathematics: the role of visualizations in math research.
Michael L. Overton, NYU Courant, Numerical Investigation of Crouzeix’s Conjecture
SAS 4201In many applications one wishes to minimize an objective function that is not convex and is not differentiable at its minimizers. We discuss two algorithms for minimization of nonsmooth, nonconvex functions. Gradient Sampling is a simple method that, although computationally intensive, has a nice convergence theory. The method is robust and the convergence theory has…
Jeff Meier, University of Georgia Athens, Filling transverse links with trisected surfaces
SAS 4201We will describe an adaptation of the theory of trisections to the setting of properly embedded, smooth, compact surfaces in smooth, compact, orientable four-manifolds with boundary. The main result is that any such surface can be isotoped to lie in bridge trisected position with respect to a given trisection on the ambient four-manifold. The trisection…
Constantine Dafermos, Brown University, Energy dissipation at maximal rate
SAS 4201The lecture will consider the case of evolution equations, endowed with a "free energy," for which the initial value problem possesses multiple solutions, with an eye to examining whether the particular solution that maximizes the rate of energy decay enjoys a special status.