Seminars and Colloquia by Series

Tuesday, January 10, 2012 - 11:00 , Location: Skiles 006 , Marisa Eisenberg , MBI, Ohio State , Organizer:
Waterborne diseases cause over 3.5 million deaths annually, with cholera alone responsible for 3-5 million cases/year and over 100,000 deaths/year. Many waterborne diseases exhibit multiple characteristic timescales or pathways of infection, which can be modeled as direct and indirect transmission.  A major public health issue for waterborne diseases involves understanding the modes of transmission in order to improve control and prevention strategies. One question of interest is: given data for an outbreak, can we determine the role and relative importance of direct vs. environmental/waterborne routes of transmission? We examine these issues by exploring the identifiability and parameter estimation of a differential equation model of waterborne disease transmission dynamics. We use a novel differential algebra approach together with several numerical approaches to examine the theoretical and practical identifiability of a waterborne disease model and establish if it is possible to determine the transmission rates from outbreak case data (i.e. whether the transmission rates are identifiable). Our results show that both direct and environmental transmission routes are identifiable, though they become practically unidentifiable with fast water dynamics. Adding measurements of pathogen shedding or water concentration can improve identifiability and allow more accurate estimation of waterborne transmission parameters, as well as the basic reproduction number. Parameter estimation for a recent outbreak in Angola suggests that both transmission routes are needed to explain the observed cholera dynamics. I will also discuss some ongoing applications to the current cholera outbreak in Haiti.
Thursday, December 8, 2011 - 11:00 , Location: Skiles 006 , Kirsten Wickelgren , AIM/Harvard University , Organizer: John Etnyre
The cohomology ring of the absolute Galois group Gal(kbar/k) of a field k  controls interesting arithmetic properties of k. The Milnor conjecture, proven by Voevodsky, identifies the cohomology ring H^*(Gal(kbar/k), Z/2) with the tensor algebra of k* mod the ideal generated by x otimes 1-x for x in k - {0,1} mod 2, and the Bloch-Kato theorem, also proven by Voevodsky, generalizes the coefficient ring Z/2. In particular, the cohomology ring of Gal(kbar/k) can be expressed in terms of addition and multiplication in the field k, despite the fact that it is difficult even to list specific elements of Gal(kbar/k). The cohomology ring is a coarser invariant than the differential graded algebra of cochains, and one can ask for an analogous description of this finer invariant, controlled by and controlling higher order cohomology operations. We show that order n Massey products of n-1 factors of x and one factor of 1-x vanish, generalizing the relation x otimes 1-x. This is done by embedding P^1 - {0,1,infinity} into its Picard variety and constructing Gal(kbar/k) equivariant maps from pi_1^et applied to this embedding to unipotent matrix groups. This also identifies Massey products of the form <1-x, x, … , x , 1-x> with f cup 1-x, where f is a certain cohomology class which arises in the description of the action of Gal(kbar/k) on pi_1^et(P^1 - {0,1,infinity}). The first part of this talk will not assume knowledge of Galois cohomology or Massey products.
Tuesday, December 6, 2011 - 11:00 , Location: Skiles 006 , Joseph Rabinoff , Harvard University , Organizer: Anton Leykin
Bernstein's theorem is a classical result which computes the number of common zeros in (C*)^n of a generic set of n Laurent polynomials in n variables. The theorem of the Newton polygon is a ubiquitous tool in arithmetic geometry which calculates the valuations of the zeros of a polynomial (or convergent power series) over a non-Archimedean field, along with the number of zeros (counted with multiplicity) with each given valuation. We will explain in what sense both theorems are very special cases of a lifting theorem in tropical intersection theory. The proof of this lifting theorem builds on results of Osserman and Payne, and uses Berkovich analytic spaces and extended tropicalizations of toric varieties in a crucial way, as well as Raynaud's theory of formal models of analytic spaces. Most of this talk will be about joint work with Brian Osserman.
Thursday, December 1, 2011 - 15:00 , Location: Skiles 005 , Dr. Jianfeng Lu , Courant Institute, NYU , Organizer: Haomin Zhou
Electronic structure theories, in particular Kohn-Sham density functional theory, are widely used in computational chemistry and material sciences nowadays. The computational cost using conventional algorithms is however expensive which limits the application to relative small systems. This calls for development of efficient algorithms to extend the first principle calculations to larger system. In this talk, we will discuss some recent progress in efficient algorithms for Kohn-Sham density functional theory. We will focus on the choice of accurate and efficient discretization for Kohn-Sham density functional theory.
Thursday, February 10, 2011 - 11:00 , Location: Skiles 006 , Svitlana Mayboroda , Purdue University , Organizer: Michael Lacey
 Despite its long history, the theory of ellipticpartial differential equations in non-smooth media is abundant with openproblems. We will discuss the main achievements in the theory, recentdevelopments, surprising paradoxes related to the behavior of solutions nearthe boundary, and some fundamental questions which still remain open.
Wednesday, February 2, 2011 - 13:00 , Location: Skiles 005 , Vladislav Kargin , Department of Mathematics, Stanford University , Organizer: Christian Houdre
Let H = A+UBU* where A and B are two N-by-N Hermitian matrices and U is a random unitary transformation. When N is large, the point measure of eigenvalues of H fluctuates near a probability measure which depends only on eigenvalues of A and B. In this talk, I will discuss this limiting measure and explain a result about convergence to the limit in a local regime.
Tuesday, February 1, 2011 - 11:00 , Location: Skiles 006 , Peter Kim , University of Utah , , Organizer:
The immune system is a complex, multi-layered biological system, making it difficult to characterize dynamically.  Perhaps, we can better understand the system’s construction by isolating critical, functional motifs.  From this perspective, we will investigate two simple, yet ubiquitous motifs:state transitions and feedback regulation.Numerous immune cells exhibit transitions from inactive to activated states.  We focus on the T cell response and develop a model of activation, expansion, and contraction.   Our study suggests that state transitions enable T cells to detect change and respond effectively to changes in antigen levels, rather than simply the presence or absence of antigen.  A key component of the system that gives rise to this change detector is initial activation of naive T cells.  The activation step creates a barrier that separates the slow dynamics of naive T cells from the fast dynamics of effector T cells, allowing the T cell population to compare short-term changes in antigen levels to long-term levels.  As a result, the T cell population responds to sudden shifts in antigen levels, even if the antigen were already present prior to the change.  This feature provides a mechanism for T cells to react to rapidly expandingsources of antigen, such as viruses, while maintaining tolerance to constant or slowly fluctuating sources of stimulation, such as healthy tissue during growth.For our second functional motif, we investigate the potential role of negative feedback in regulating a primary T cell response.  Several theories exist concerning the regulation of primary T cell responses, the most prevalent being that T cells follow developmental programs.  We propose an alternative hypothesis that the response is governed by a feedback loop between conventional and adaptive regulatory T cells.  By developing a mathematical model, we show that the regulated response is robust to a variety of parameters and propose that T cell responses may be governed by a simple feedback loop rather than by autonomous cellular programs.
Wednesday, January 26, 2011 - 15:00 , Location: Skiles 006 , Aaron Lauda , Columbia University , Organizer: Stavros Garoufalidis
The Jones polynomial is a link invariant that can be understood in terms of the representation theory of the quantum group associated to sl2. This description facilitated a vast generalization of the Jones polynomial to other quantum link and tangle invariants called Reshetikhin-Turaev invariants. These invariants, which arise from representations of quantum groups associated to simple Lie algebras, subsequently led to the definition of quantum 3-manifold invariants. In this talk we categorify quantum groups using a simple diagrammatic calculus that requires no previous knowledge of quantum groups. These diagrammatically categorified quantum groups not only lead to a representation theoretic explanation of Khovanov homology but also inspired Webster's recent work categorifying all Reshetikhin-Turaev invariants of tangles.
Monday, January 24, 2011 - 15:00 , Location: Skiles 006 , Greg Blekherman , University of California, San Diego , Organizer: John Etnyre
A multivariate real polynomial $p$ is nonnegative if $p(x) \geq 0$ for all $x \in R^n$. I will review the history and motivation behind the problem of representing nonnegative polynomials as sums of squares. Such representations are of interest for both theoretical and practical computational reasons. I will present two approaches to studying the differences between nonnegative polynomials and sums of squares. Using techniques from convex geometry we can conclude that if the degree is fixed and the number of variables grows, then asymptotically there are significantly more nonnegative polynomials than sums of squares. For the smallest cases where there exist nonnegative polynomials that are not sums of squares, I will present a complete classification of the differences between these sets based on algebraic geometry techniques.
Thursday, January 20, 2011 - 15:00 , Location: Skiles 005 , Jelena Bradic , Princeton University , Organizer: Liang Peng
High throughput genetic sequencing arrays with thousands of measurements per sample and a great amount of related censored clinical data have increased demanding need for better measurement specific model selection. In this paper we establish strong oracle properties of non-concave penalized methods for non-polynomial (NP) dimensional data with censoring in the framework of Cox's proportional hazards model. A class of folded-concave penalties are employed and both LASSO and SCAD are discussed specifically. We unveil the question under which dimensionality and correlation restrictions can an oracle estimator be constructed and grasped. It is demonstrated that non-concave penalties lead to significant reduction of the "irrepresentable condition" needed for LASSO model selection consistency. The large deviation result for martingales, bearing interests of its own, is developed for characterizing the strong oracle property. Moreover, the non-concave regularized estimator, is shown to achieve asymptotically the information bound of the oracle estimator. A coordinate-wise algorithm is developed for finding the grid of solution paths for penalized hazard regression problems, and its performance is evaluated on simulated and gene association study examples.