Professor Guo Lin
Lanzhou University, China

Traveling Wave Solutions of Evolutionary Models without Monotonicity

Monday, May 9, 2016, 4:00pm
Ungar Room 402

Abstract: This talk is concerned with the traveling wave solutions of evolutionary systems including delayed reaction-diffusion systems and integrodifference equations. Even if the general monotone conditions fail, the existence of traveling wave solutions is studied by generalized upper and lower solutions. The asymptotic behavior is established by the idea of contracting rectangles. In the study of classical Lotka-Volterra competitive systems, we obtain the existence of nonmonotone traveling wave solutions, which weakly confirms the conjecture by Tang and Fife [ARMA, 1980].

Dr. Dmitri Vassiliev
University College London

Spectral Asymptotics for First Order Systems

Thursday, April 14, 2016, 5:00pm
Ungar Room 402

Abstract: In layman's terms a typical problem in this subject area is formulated as follows. Suppose that our universe has finite size but does not have a boundary. An example of such a situation would be a universe in the shape of a 3-dimensional sphere embedded in 4-dimensional Euclidean space. And imagine now that there is only one particle living in this universe, say, a massless neutrino. Then one can address a number of mathematical questions. How does the neutrino field (solution of the massless Dirac equation) propagate as a function of time? What are the eigenvalues (energy levels) of the particle? Are there nontrivial (i.e. without obvious symmetries) special cases when the eigenvalues can be evaluated explicitly? What is the difference between the neutrino (positive energy) and the antineutrino (negative energy)? What is the nature of spin? Why do neutrinos propagate with the speed of light? Why are neutrinos and photons (solutions of the Maxwell system) so different and, yet, so similar?

The speaker will approach the study of first order systems of partial differential equations from the perspective of a spectral theorist using techniques of microlocal analysis and without involving geometry or physics. However, a fascinating feature of the subject is that this purely analytic approach inevitably leads to differential geometric constructions with a strong theoretical physics flavour.

Alexander Volberg
Michigan State University

Rectifiability of Harmonic Measure

Wednesday, April 13, 2016, 4:00pm
Ungar Room 402

Abstract: In a recent multi-authored paper by J. Azzam, S. Hofmann, J.-M. Martell, S. Mayboroda, M. Mourgoglou, X. Tolsa and myself the following result is proved:

For arbitrary open set in any \R^d such that its boundary carries harmonic measure w and for any Borel set $E$, such that $H^{d-1}(E)<\infty, if harmonic measure w|E is absolutely continuous with respect to H^{d-1}|E, then it is rectifiable.

This result solves a long standing conjecture of Chris Bishop and generalizes the result of N. Makarov obtained for d=2 and for simply connected open sets obtained in the early 80's.

We will also make an overview of harmonic measure results of the last 30 years.

Weiyi Zhang
University of Warwick

J-holomorphic Subvarieties

Friday, March 18, 2016, 5:45pm
Ungar Room 402

Abstract: In this talk, we discuss the J-holomorphic subvarieties in a 4-dimensional symplectic manifold. We will start by showing that a subvariety of a complex rational surface in an exceptional rational curve class could have higher genus components. On the other hand, such exotic phenomenon won't happen for (any tamed almost complex structures on) ruled surfaces.

We will then show a general cone theorem as in Mori theory and determine the moduli space of a spherical class in a rational or ruled surface, which could be decomposed as linear systems.

Professor Fedor Bogomolov
New York University

Comparing Torsion Points of Elliptic Curves

Friday, March 18, 2016, 4:45pm
Ungar Room 402

Abstract: In this talk we consider subset of $P^1_{E}\subset P^1$ which are the images of torsion points of complex elliptic curve $E$ under projection defined as a quotient map for standard involution. The corresponding map $E\to P^1$ and hence $P^1_{E}\subset P^1$ is uniquely defined modulo projective transform on $P^1$. It is known that the intersection of subsets $P^1_{E}, P^1_{E'}$ are finite apart from the case when they coincide. However the question is whether there is universal constant bounding this intersection. This problem is somewhat related to Serre conjecture on the image of Galois action on the torsion points. I am going to discuss some results concerning the matter. In particular we show that the intersection contains exactly one point for some families of elliptic curves, that there is an infinite set of pairs $E,E'$ where the intersection consists of $10$ points and also provide a example with $14$ intersection points.

This is joint work with Hang Fu.

Dr. Phillip Griffiths
Institute of Advanced Study, Princeton
University of Miami

The Hodge Conjecture

Friday, February 12, 2015, 5:00pm
Ungar Room 402

Abstract: The Hodge conjecture is one of the Millennium Prize problems. In this talk I will discuss the questions:

•What is it?
•Why is it interesting?
•What is known about it?
•Why has there been almost no progress in solving it?

As little background as possible will be assumed, and what is needed will be informally explained.

Based on joint work with Mark Green, Radu Laza and Colleen Robles.

Ilya Kachkovskiy
University of California, Irvine

Almost Commuting Operators

Tuesday, January 26, 2016, 5:00pm
Ungar Room 402

Abstract: The classical question of whether a pair of matrices or operators with small commutator is close to a commuting pair dates back to von Neumann, Rosenthal, and Halmos. In particular, a dimension-uniform version of this question for Hermitian matrices was open until 1993 when Lin proved the existence of a uniform distance estimate using deep C*-algebraic arguments. Lin's method didn't allow to obtain any quantitative results besides existence. In 2001, Davidson and Szarek conjectured that the distance is bounded by some universal constant times the square root of the norm of the commutator.

In this talk, I will explain the main ideas of the proof of Davidson and Szarek's conjecture (joint work with Y. Safarov), including the infinite-dimensional version.

I will also talk about the relation with Brown-Douglas-Fillmore theorem, and the case of the Hilbert-Schmidt norm for matrices. If time permits, I will discuss some work in progress that includes possible generalization to the case of unitary matrices (where an additional topological obstruction arises), and some applications to mathematical physics.

Dr. Antoine Choffrut
University of Warwick

The Analysis of Partial Differential Equations:
A Case Study with the Euler Equations of Fluid Mechanics

Thursday, January 21, 2016, 5:00pm
Ungar Room 402

Abstract: In this talk I will give an overview of my work on the incompressible Euler equations. It can be categorized in two classes: the use of convex integration to prove h-principles, and structure theorems for two-dimensional stationary flows. These are two very different types of results. However, I will try to highlight a unifying principle in this body of work.

Mihaela Ifrim
University of California, Berkeley

Two Dimensional Water Waves in Holomorphic Coordinates

Thursday, January 14, 2016, 5:00pm
Ungar Room 402

Abstract: This is joint work with Daniel Tataru, and in parts with Benjamin Harrop-Griffits and John Hunter. My talk is concerned with the infinite depth water wave equation in two space dimensions, with either gravity or surface tension. I will also make some remarks on the finite depth case, and on the infinite depth case in which constant vorticity and only gravity are assumed. We consider this problem expressed in position-velocity potential holomorphic coordinates. Viewing this problem(s) as a quasilinear dispersive equation, we develop new methods which will be used to prove enhanced lifespan of solutions and also global solutions for small and localized data. For the gravity water waves there are several results available; they have been recently obtained by Wu, Alazard-Burq-Zuily and Ionescu-Pusateri using different coordinates and methods. In the capillary water waves case, we were the first to establish a global result (two months later, Ionescu-Pusateri also announced a related result). Our goal is improve the understanding of these problems by providing a single setting for all the above cases, and presenting simpler proofs. The talk will try to be self contained.

Mimi Dai
University of Illinois at Chicago

Regularity for the 3D Navier-Stokes Equations and Related Problems

Tuesday, January 12, 2016, 5:00pm
Ungar Room 402

Abstract: As one of the most significant problems in the study of partial differential equations arising in fluid dynamics, Leray's conjecture in 1930's regarding the appearance of singularities for the 3-dimensional (3D) Navier-Stokes equations (NSE) has been neither proved nor disproved. The problems of blow-up have been extensively studied for decades using different techniques. By using a method of wavenumber splitting which originated from Kolmogorov's theory of turbulence, we obtained a new regularity criterion for the 3D NSE. The new criterion improves the classical Prodi-Serrin, Beale-Kato-Majda criteria and their extensions. Related problems, such as the well/ill-posedness, will be discussed as well.

Dr. Eduardo Sontag
Rutgers University

Qualitative Features of Transient Responses
A Case Study: Scale-invariance

Thursday, December 17, 2015, 4:00pm
Ungar Room 402

Abstract: An ubiquitous property of sensory systems is "adaptation": a step increase in stimulus triggers an initial change in a biochemical or physiological response, followed by a more gradual relaxation toward a basal, pre-stimulus level. Adaptation helps maintain essential variables within acceptable bounds and allows organisms to readjust themselves to an optimum and non-saturating sensitivity range when faced with a prolonged change in their environment. It has been recently observed that some adapting systems, ranging from bacterial chemotaxis pathways to signal transduction mechanisms in eukaryotes, enjoy a remarkable additional feature: scale invariance or "fold change detection" meaning that the initial, transient behavior remains approximately the same even when the background signal level is scaled.

I will review the biological phenomenon, and formulate a theoretical framework leading to a general theorem characterizing scale invariant behavior by equivariant actions on sets of vector fields that satisfy appropriate Lie-algebraic non-degeneracy conditions. The theorem allows one to make experimentally testable predictions, and I will discuss the validation of these predictions using genetically engineered bacteria and microfluidic devices, as well their use as a "dynamical phenotype" for model invalidation. I will conclude by briefly engaging in some wild and irresponsible speculation about the role of the shape of transient responses in immune system self/other recognition and in evaluating the initial effects of immunotherapy.

Dr. Phillip Griffiths
Institute of Advanced Study, Princeton

Hodge Theory and Moduli

Wednesday, September 9, 2015, 5:00pm
Ungar Room 402

Abstract: This talk will give an overview of the connection between the above two topics in the first non-classical case of H-surfaces.

Based on joint work with Mark Green, Radu Laza and Colleen Robles.

Professor Wayne Lawton
University of Western Australia

Multivariate Prediction and Spectral Factorization

Thursday, September 3, 2015, 5:00pm
Ungar Room 402

Abstract: Prediction theory studies stationary random functions on ordered groups. Time series are functions on the integer group and characterized by classical harmonic analysis results such as Szego's spectral factorization theorem. Images are functions on higher rank groups. We use results about entire functions and ergodic theory to derive new results for these functions.

Dr. Dimitri Gurevich
Valenciennes University, France

From Quantum Groups to Noncommutative Geometry

Tuesday, May 5, 2015, 4:00pm
Ungar Room 402

Abstract: Since creation theory of Quantum Groups numerous attempts to elaborate an appropriate differential calculus were undertaken. Recently, a new type of Noncommutative Geometry has been obtained on this way. Namely, we have succeeded in introducing the notions of partial derivatives on the enveloping algebras U(gl(m)) and constructing the corresponding de Rham complexes. All objects arising in our approach are deformations of their classical counterparts. In my talk I plan to introduce some basic notions of the theory of Quantum Groups and to exhibit possible applications of thistype Noncommutative Geometry to quantization of certain dynamical models.

Dr. Phillip Griffiths
Institute for Advanced Study, Princeton

Hodge Theory: Some History and a Possible Future Direction

Tuesday, April 21, 2015, 5:00pm
Ungar Room 402

Abstract: Hodge theory is a central part of algebraic geometry. The subject has a long and rich history and continues today to be at the forefront of much current work. This talk will discuss some of its history and one possible, less well known, possible area for future work. In brief: where did the subject come from, and where might it be heading?

Professor Yuval Roichman
Bar-Ilan University

Fine Sets and Schur Positivity

Tuesday, April 14, 2015, 5:00pm
Ungar Room 402

Abstract: Characters of symmetric groups and Iwahori-Hecke algebras may be evaluated by signed (and slightly weighted) enumerations of various classes of permutations, such as conjugacy classes, Knuth classes and inverse descent classes. We propose an abstract framework for such classes, which we call "fine sets".

It will be shown that fine sets can be characterized by Schur positivity of the associated quasi-symmetric functions. The proof involves asymmetric Walsh-Hadamard type matrices. Applications include the equivalence of classical theorems of Lusztig-Stanley and Foata-Schutzenberger.

Time allowing, some very recent developments will be described -- in particular, a generalization to type B and applications to grid classes and derangements, related to results of Desarmenien-Wachs and others.

Based on joint works with R. Adin, C. Athanasiadis and S. Elizalde.

Professor Alexei Kovalev
University of Cambridge

8-dimensional Manifolds with Holonomy Spin(7)

Monday, April 6, 2015, 4:00pm
Ungar Room 402

Abstract: Spin(7) is one of the "exceptional" holonomy groups of Riemannian metrics appearing in Berger's classification. Riemannian manifolds with holonomy Spin(7) occur in dimension 8, are Ricci-flat and have parallel spinor fields. We construct examples of asymptotically cylindrical Riemannian 8-manifolds with holonomy group Spin(7). To our knowledge, these are the first such examples. This leads to a new, connected sum construction of compact holonomy-Spin(7) manifolds from asymptotically cylindrical pairs. We show in examples that the construction produces a "pulling-apart" deformation of compact Spin(7)-manifolds previously given by Joyce, as well as topologically new Spin(7)-manifolds.

Professor Xiuxiong Chen
Stony Brook University

On the Kaehler Ricci Flow

Friday, March 27, 2015, 5:00pm
Ungar Room 402

Abstract: There is a long standing conjecture on Kaehler Ricci flow on Fano manifolds that the Ricci flow converges sub-sequentially to a Kaehler Ricci solution with at most codimension 4 singularities, with perhaps a different complex structure (the so-called "Hamilton-Tian conjecture"). In this lecture, we will outline a proof of this conjecture. This is a joint work with Bing Wang.

Professor Mario Milman
Florida Atlantic University

Uncertainty Inequalities in Metric Measure Spaces

Friday, March 20, 2015, 5:00pm
Ungar Room 402

Abstract: I will show an extension of the classical uncertainty inequalities to connected metric measure spaces. The idea is to exploit the connection with isoperimetry. I will introduce a new class of weights (*isoperimetric weights*) and prove a new local Poincare inequality formulated in terms of the isoperimetric profile. A reinterpretation of the notion of *isoperimetric weights* in terms of function spaces provides a different method to attack the problem and also allows for explicit computation of these weights for log concave measures and other model geometries. I will try to explain all the terms in the abstract, provide examples and show connections with classical inequalities by Hardy and Strichartz.

Professor King-Leung Lam
Ohio State University

On Global Dynamics of Competitive Systems in Ordered Banach Spaces

Tuesday, March 17, 2015, 5:00pm
Ungar Room 402

Abstract: A well-known result in [Hsu-Smith-Waltman, Trans. AMS (1996)] states that in a competitive semiflow defined on the product of two cones in respective Banach spaces, one of the following outcomes is possible for the two competitors: either there is at least one stable coexistence steady state, or else one of the exclusion states attracts all trajectories initiating in the order interval bounded by the two exclusion states. However, none of the exclusion states can be globally asymptotically stable if we broaden our scope to the entire positive cone. In this talk, we discuss two sufficient conditions that guarantee, in the absence of coexistence steady states, the global asymptotic stability of one of the exclusions states. Our results complement the counter example mentioned in the above paper and are frequently applicable in practice.

Dr. Alexander Volberg
Michigan State University

Why the Oracle May Not Exist:
Ergodic Families of Jacobi Matrices,
Absolute Continuity without Almost Periodicity

Thursday, March 5, 2015, 5:00pm
Ungar Room 402

Abstract: We will explain the recent solution of Kotani's problem pertinent to the existence/non-existence of "oracle" (almost periodicity) for the ergodic families of Jacobi matrices (discrete Schroedinger operators). Kotani suggested that such families are subject to the following implication: if family has a non-trivial absolutely continuous spectrum (this happens almost surely) then almost surely it consists of almost periodic matrices (hence the possibility to predict the future by the past). Kotani proved an important positive result of this sort. Recently independently Artur Avila and Peter Yuditskii-myself disproved this conjecture of Kotani (by two different approaches). We will show the hidden singularity that defines when such Kotani's oracle exists or not.

S.R.S. Varadhan
Frank J. Gould Professor of Science
Courant Institute at New York University

Large Deviations for Brownian Occupation Times, Revisited

Friday, February 20, 2015, 5:00pm
Ungar Room 402

Abstract: Brownian motion on R^d is not positive recurrent and the Large Deviation estimate holds only in a weak form. To get a strong version a compactification is needed. In the application of interest we need a translation invariant compactification and that is carried out.

Dr. Lingjiong Zhu
University of Minnesota

Self-Exciting Point Processes

Tuesday, February 17, 2015, 5:00pm
Ungar Room 402

Abstract: Self-exciting point processes are simple point processes that have been widely used in neuroscience, sociology, finance and many other fields. In many contexts, self-exciting point processes can model the complex systems in the real world better than the standard Poisson processes. We will discuss the Hawkes process, the most studied self-exciting point process in the literature. We will talk about the limit theorems and asymptotics in different regimes. Extensions to Hawkes processes and applications to finance will also be discussed.

Dr. Karim Adiprasito
Institute for Advanced Study

Toric Chordality

Monday, February 16, 2015, 5:00pm
Ungar Room 402

Abstract: We put the fundamental graph-theoretic notion of chordality into a proper context within the weight algebra of McMullen and framework rigidity. Moreover, we generalize some of the classical results of graph chordality to this context, including the fundamental relation to the Leray property.

Our main focus is the relation of higher chordality to the Hard-Lefschetz Theorem for simplicial polytopes of Saito, McMullen and others. Homological chordality allows us to state a powerful Quantitative Lower Bound Theorem which relates the "defect" to a chordal complex to the g-numbers of the same polytope, thereby providing an immediate combinatorial interpretation of the g-numbers and, more refinedly, the cohomology classes of the associated projective toric variety, in terms of induced subcomplexes of the simplicial complex. While most of our results follow quite easily once we established stress spaces as the right setting for the study of higher chordality, we provide a central propagation theorem for toric chordality. As an application, we resolve a variety of interesting questions in combinatorics of polytopes.

Dr. Hamed Amini
Swiss Federal Institute of Technology (EPFL), Lausanne

Systemic Risk in Financial Networks

Thursday, February 12, 2015, 5:00pm
Ungar Room 402

Abstract: After the last financial crisis, the monitoring and managing of systemic risk has emerged as one of the most important concerns for regulators, governments and market participants. In this talk, we propose two complementary approaches to modeling systemic risk.

In the first approach, we model the propagation of balance-sheet or cash-flow insolvency across financial institutions as a cascade process on a network representing their mutual exposures. We derive rigorous asymptotic results for the magnitude of contagion in a large financial network and give an analytical expression for the asymptotic fraction of defaults, in terms of network characteristics. We also introduce a criterion for the resilience of a large inhomogeneous financial network to initial shocks that can be used as a tool for monitoring systemic risk. Using an equilibrium approach, in the second part of the talk, we apply a coherent systemic risk measure to examine the effects on systemic risk and liquidation losses of multilateral clearing via a central clearing counterparty (CCP). We provide sufficient conditions in terms of the CCP's fee and guarantee fund policy for a reduction of systemic risk.

Dr. Jin Hyuk Choi
Carnegie Mellon University

Taylor Approximation of Incomplete Radner Equilibrium Models

Tuesday, February 10, 2015, 5:00pm
Ungar Room 402

Abstract: We will first explain the notion of Radner equilibria by using the one-period binomial model. We will see how the complete market assumption simplifies the mathematical structure. In the next part we will consider the setting of exponential investors and uncertainty governed by Brownian motions. Here we will show the existence of an equilibrium for a general class of incomplete models. Finally, we will show that the general incomplete equilibrium can be approximated by a tractable equilibrium stemming from exponential-quadratic models.

Dr. Volker Schlue
University of Toronto

Non-existence of Time-periodic Dynamics in General Relativity

Monday, February 2, 2015, 5:00pm
Ungar Room 402

Abstract: General Relativity presents us with a theory for the dynamics of space and time itself. In this lecture I will revisit the expectation that in contrast to Newton's theory of gravity there does not exist time-periodic motion. In particular, I will present a recent result that establishes purely on the basis of knowing the gravitational waves emitted from any self-gravitating system, that any time-periodic space-time is necessarily "stationary", that is time-independent. Moreover, I will elaborate on novel unique continuation results for hyperbolic partial differential equations that play an important role in the proof, and have been obtained jointly with Spyros Alexakis and Arick Shao.

Dr. Alejandro Morales
University of California, Los Angeles

Matrices over Finite Fields, Polytopes, and Symmetric Functions Related to Rook Placements

Wednesday, January 28, 2015, 5:00pm
Ungar Room 402

Abstract: Permutation matrices are fundamental objects in combinatorics and algebra. Such matrices can be viewed as placements of non-attacking rooks on a square board. If we restrict the support of the matrix we obtain rook placements that in computational complexity are an example of problems where finding a solution is easy but the enumeration is difficult. In this talk we look at various recent results where rook placements appear.

The first instance is counting invertible matrices over a finite field of size q with support that avoids some entries. The number of such matrices is a q-analogue of rook placements but this number may not be a polynomial in q (Stembridge). We give polynomial formulas when the forced zeroes are in the diagonal and on the inversions of some permutations. We also relate the latter to the zoo of combinatorial objects in Postnikov's study of the positive Grassmannian. (Joint works with A. Klein, J. Lewis, R. Liu, G. Panova, S. V Sam, Y. Zhang.)

The second instance is as vertices of the Chan-Robbins-Yuen polytope, a face of the Birkhoff polytope of doubly stochastic matrices. This polytope has a remarkable volume formula proved by Zeilberger. We discuss three variants of this polytope. (Joint works with K. Meszaros, B. Rhoades and J. Striker.)

The third instance is as coefficients in relations among Stanley's chromatic symmetric function and Shareshian-Wachs chromatic quasisymmetric functions on a Catalan family of graphs. (Ongoing work with M. Guay-Paquet and F. Saliola.)

Dr. Antoine Choffrut
University of Edinburgh

Rigidity and Flexibility for the Euler Equations:
The h-principle and the Method of Convex Integration

Tuesday, January 27, 2015, 5:00pm
Ungar Room 402

Abstract: Certain problems in Differential Geometry have long been observed to feature a dichotomy "rigidity vs. flexibility". Rigidity refers to the fact that smooth solutions are unique, while flexibility means that the set of solutions with low regularity can be extremely large. With the groundbreaking work of De Lellis and Szekelyhidi on the incompressible Euler equations of fluid mechanics, it came as an enormous surprise that such a dichotomy may also hold for problems from Mathematical Physics. Flexibility results of interest to us fall under the name of h-principles and are established using the method of convex integration. In this talk I will give a bird's eye view of these notions and explain a motivating conjecture due to Onsager. I will conclude with future problems which deserve special research efforts.

Dr. Angela Hicks
Stanford University

Combining the Classical and the Rational:
Re-expressing Rational Shuffle Conjecture

Friday, January 23, 2015, 5:00pm
Ungar Room 402

Abstract: Recent results have place the classical shuffle conjecture in the context of an infinite family of conjectures about parking functions in any rectangular lattice. The classical case describes three traditional statistics which have varying degrees of complication: area, dinv, and pmaj. Combining the work of many authors in several fields gives two of these statistics in the more general case. After introducing the conjectures, this talk will cover how these statistics can be expressed in a manner consistent with the current classical literature and why this is important. Additionally, this talk will include a conjectured generalization of the third statistic, pmaj, in a number of new cases and time permitting, explain its connection to the sandpile model.

Dr. Nicolas Addington
Duke University

Recent Developments in Rationality of Cubic 4-folds

Tuesday, January 20, 2015, 5:00pm
Ungar Room 402

Abstract: The question of which cubic 4-folds are rational is one of the foremost open problems in algebraic geometry. I'll start by explaining what this means and why it's interesting; then I'll discuss three approaches to solving it (including one developed in the last year), my own work relating the three approaches to one another, and the troubles that have befallen each approach.

Dr. Kristin Shaw
Technical University of Berlin

Tropical Curves and Surfaces

Friday, January 16, 2015, 5:00pm
Ungar Room 402

Abstract: Tropical geometry is a relatively new field of mathematics which provides us with piecewise linear models for classical geometry. Often these models encode properties of the original objects in a more combinatorial form. The first major application of the field to classical geometry came by way of Mikhalkin's correspondence theorem between planar tropical and complex curves. This correspondence has served to answer questions in enumerative geometry; how many curves of a given degree and genus satisfy an appropriate set of constraints? Various other applications have been to moduli problems, linear series on curves, intersection theory, and computations of Hodge numbers. Tropical geometry has also been suggested as a explanation for mirror symmetry.

Abstract tropical varieties in higher dimensions have not been studied to the extend of curves. Applications of tropical geometry to classical geometry often rely on establishing correspondences between the two worlds. In this talk we will see that upon leaving the plane for other surfaces or varieties of higher dimension these correspondences are harder to establish and sometimes do not exist. We will explain how intersection theory and tropical homology can be used to study curves and also tropical Picard groups in surfaces and show how operations known as tropical modification and summation can construct and help study interesting examples of tropical surfaces.

Dr. Morgan Brown
University of Michigan

Rationality of Algebraic Varieties

Thursday, January 15, 2015, 5:00pm
Ungar Room 402

Abstract: An algebraic variety is called rational if it has a parametrization by rational functions which is one-to-one almost everywhere. Determining whether a variety is rational or not is an old and very difficult problem, even when that variety is a smooth hypersurface in projective space. I will give a brief history of this problem, and explain some of the modern techniques for showing that varieties are irrational, including how certain aspects of the derived category might reflect rationality.

Dr. Bruno Benedetti
Free University of Berlin

Diameter of Polytope Graphs

Tuesday, January 13, 2015, 5:00pm
Ungar Room 402

Abstract: "Discrete" geometry focuses on geometric objects that have facets, corners, vertices, rather than on objects with a uniform smooth structure. A standard result is Balinski's theorem, "the graph of every convex d-dimensional polytope is d-connected". But given a d-polytope with n facets, how many edges do we have to walk along (at most), if we want to go from a vertex to another? This "combinatorial diameter" question has relevance in optimization, being related to the simplex algorithm. Hirsch conjectured in 1957 that the answer should be "at most n-d", but the conjecture was disproved by Santos in 2010. I will explain two recent positive results:

(1) The Hirsch conjecture holds for all flag polytopes. The proof uses methods from differential geometry. (Joint work with K. Adiprasito.)

(2) The notion of dual graph naturally extends to projective curves, or to arbitrary algebraic varieties. Surprisingly, a broader version of Balinski's theorem still holds. (Joint work with M. Varbaro, and work in progress with B. Bolognese.)

Dr. Xing Liang
Chinese University of Science and Technology

Spreading Speeds of Monostable Parabolic Equations in Periodic Habitat with Free Boundary

Wednesday, December 17, 2014, 4:30pm
Ungar Room 402

Abstract: In this talk, we will show the existence of the speading speeds of monostable parabolic equations in the spatially and temporally periodic habitats. We main focus on the case where the equations is not linearly determinate.

Dr. James Keesling
Department of Mathematics
University of Florida

An Agent Based Microsimulation Model for the Spread of Citrus Greening

Thursday, November 13, 2014, 5:00pm
Ungar Room 402

Abstract: The citrus industry is a pillar of Florida's economy. It is hard to imagine that it may disappear in short order. However, due to the devastating effects of Citrus Greening, this may very well be the case.

Citrus Greening (or Huang Long Bing) is a disease of citrus caused by a bacterium (Candidatus Liberibacter asiaticus). It is spread by a psyllid, Diaphorina citri. In this talk we will describe a simulation model for the spread of this disease. The model is based on a detailed analysis of the biology of the psyllid vector. Current work expanding and applying the model is being funded by the Citrus Research and Development Foundation as part of the Psyllid Shield Project.

Professor John Stillwell
University of San Francisco

What Does "Depth" Mean in Math?
History, Foundations, and Logic

Thursday, November 6, 2014, 5:00pm
Witten Learning Center 160

Abstract: Every mathematician believes that certain theorems are "deep," but the concept of depth does not have a formal definition. By looking at some famous theorems, ancient and modern, we will study some candidates for "depth" at various levels, particularly the undergraduate level. With these examples in hand we hope to discuss whether any concepts of logic now available can give "depth" a precise meaning.

View Poster

Anatoly Libgober
University of Illinois at Chicago

Landau-Ginzburg/Calabi-Yau and McKay Correspondences for Elliptic Genus

Monday, October 27, 2014, 5:00pm
Ungar Room 506

Abstract: I will discuss elliptic genus of singular varieties and its extension to Witten's phases of N=2 theories. In particular McKay correspondence for elliptic genus will be described. As one of applications I will show how to derive relations between elliptic genera of Calabi-Yau manifolds and related Witten phases using equivariant McKay correspondence for elliptic genus.

Dr. Vyacheslav Shokurov
Johns Hopkins University

Weak Kawamata Conjecture

Wednesday, October 15, 2014, 5:00pm
Ungar Room 506

Abstract: The talk will explain the finiteness of weak log canonical models related by bounded flops.

Dr. Kimihiko Motegi
Nihon University, Tokyo, Japan

Twisted Families of L-space Knots

Wednesday, September 10, 2014, 5:00pm
Ungar Room 402

Abstract: A knot in the 3-sphere is called an L-space knot if it admits a nontrivial Dehn surgery yielding an L-space, which is a generalization of a lens space from the algebraic viewpoint of Heegaard Floer homology. Given an L-space knot K, can we obtain an infinite family of L-space knots by twistings of K along a suitably chosen unknotted circle? We will discuss this question in the case where K admits a Seifert surgery, and give a sufficient condition on such an unknotted circle. If K is a torus knot, then we have an unknotted circle c such that twistings along c produce an infinite family of hyperbolic, L-space knots. In particular, for the trivial knot we can take infinitely many such unknotted circles. We also demonstrate that there are infinitely many hyperbolic, L-space knots with tunnel number greater than one, each of which arises from a trefoil knot by alternate twistings along two unknotted circles.

Dr. Ryan Hynd
Department of Mathematics
University of Pennsylvania

Option Pricing in the Large Risk-aversion, Small-transaction-cost Limit

Thursday, August 21, 2014, 4:00pm
Ungar Room 402

Abstract: We discuss an alternative to the well-known Black-Scholes option pricing model and characterize the asking price of a risk-averse seller, in the limit of small transaction costs. The resulting mathematical problem is one of asymptotic analysis of partial differential equations and the real challenge involves a particular eigenvalue problem.

Professor Andrea Nahmod
University of Massachusetts, Amherst

Randomization and Long Time Dynamics in Nonlinear Evolution PDE

Tuesday, April 29, 2014, 5:00pm
Ungar Room 402

Abstract: The field of nonlinear dispersive equations has undergone significant progress in the last twenty years thanks to the influx of tools and ideas from nonlinear Fourier and harmonic analysis, geometry and analytic number theory, into the existing functional analytic methods. This body of work has primarily focused on deterministic aspects of wave phenomena and answered fundamental questions such as existence and long time behavior of solutions, in various regimes. Yet there remain some important obstacles and open questions.

A natural approach to tackle some of them, and one which has recently seen a growing interest, is to consider certain evolution equations from a non-deterministic point of view (e.g. the random data Cauchy problem, invariant measures, etc.) and incorporate to the deterministic toolbox, powerful but still classical tools from probability as well.

Such approach goes back to seminal work by Bourgain in the mid 90's where global well-posedness of certain periodic Hamiltonian PDEs was studied in the almost sure sense via the existence and invariance of their associated Gibbs measures.

In this talk we will explain these ideas, describe some recent work and future directions with an emphasis on the interplay of deterministic and probabilistic approaches.

Professor Ian Hambleton
McMaster University

Manifolds and Symmetry

Wednesday, April 23, 2014, 5:00pm
Ungar Room 402

Abstract: This will be survey talk about connections between the topology of a manifold and its group of symmetries. I will illustrate this theme by discussing finite group actions on compact manifolds, such as spheres or products of spheres, and infinite discrete groups acting properly discontinuously on non-compact manifolds, such as products of spheres and Euclidean spaces.

Dr. Andras Nemethi
Hungarian Academy of Sciences

Lattice Cohomology and the Geometric Genus of Surface Singularities

Monday, April 14, 2014, 5:00pm
Ungar Room 402

Abstract: The link of a complex normal surface singularity is an oriented 3-manifold. We will review the definition of the lattice cohomology associated with such a plumbed 3-manifold. It makes connection between low-dimensional topology and singularity theory. E.g., its Euler characteristic is the Seiberg-Witten invariant of the link, while in 'nice' cases it also provides the geometric genus of the analytic structure. We will focus on such connections and interplays, together with a historical overview of the topological characterizations of the geometric genus.

Professor Fedor Bogomolov
New York University

On Homomorphisms between Multiplicative Groups of Fields

Wednesday, March 19, 2014, 5:00pm
Ungar Room 402

Abstract: I am going to describe a proof of the theorem which allows for any such homomorphism which respects algebraic dependence between elements is related to a nonarchimedean valuation of the initial field. The only additional condition is that the homomorphism has at least two algebraically independent elements in the image and a nontrivial kernel (not equal to a subfield). This result can be considered as rational version of so called Grothendieck section conjecture for functional fields of transcendence degree greater or equal to two.

Dr. John Shareshian
Washington University

Coset Posets and Probabalistic Zeta Functions of Finite Groups, and a Problem on Binomial Coefficients

Monday, March 17, 2014, 5:00pm
Ungar Room 402

Abstract: Philip Hall introduced the general Möbius inversion formula in order to enumerate the set of k-tuples (g1,...,gk) from a finite group G that contain a generating set for G. One can substitute an arbitrary complex number, rather than just a positive integer k into the formula obtained by Hall, and the resulting function is called the probabalistic zeta function for G. Serge Bouc observed that evaluation of the probabalistic zeta function at -1 yields the reduced Euler characteristic of a topological space naturally associated to G, namely, the order complex of the poset of all cosets of all proper subgroups of G. Ken Brown investigated this complex and was led to ask whether it is ever contractible. In joint work with Russ Woodroofe, we show that this complex has nontrivial homology, and is therefore not contractible, whenever G has no alternating group as a composition factor. Our efforts to understand alternating composition factors led to a fascinating (at least to me) elementary problem about prime divisors of binomial coefficients.

Dr. Priyanga Amarasekare

Temperature Effects on Population Dynamics and Species Interactions:
A Trait-based Perspective

Friday, March 7, 2014, 5:00pm
Ungar Room 402

Abstract: Populations and communities are complex systems whose properties result from the interplay between non-linear feedbacks that are intrinsic to the system (e.g., biotic interactions that lead to density- and frequency-dependence) and external inputs (e.g., abiotic factors) that are outside the feedback structure of the system. Understanding this interplay requires that we understand the mechanisms by which the effects of external inputs on lower levels of the system (e.g., traits of organisms) influence properties at higher levels (e.g., population viability, species diversity). Using temperature as the axis of abiotic variation, I develop a mechanistic theoretical framework for elucidating how abiotic effects on traits influence population dynamics and species interactions, and how these ecological dynamics in turn feedback into the trait response, causing trait evolution. I test model predictions with data on insects. The integration of theory and data paves the way for making testable predictions about the effects of climate warming on population viability, biodiversity and the control of invasive species.

Dr. Gustavo Ponce
UC Santa Barbara

On Unique Continuation Properties of Solutions to Some Dispersive Equations

Thursday, March 6, 2014, 5:00pm
Ungar Room 402

Abstract: We shall discuss results concerning unique continuationproperties of solutions to some canonical dispersive equations and their relation with decay and persistent properties of the corresponding solutions flow. These canonical dispersive models include the generalized Korteweg-de Vries equation, the nonlinear Schrödinger equation and the Benjamin-Ono.

Dr. Emil Wiedemann
PIMS Postdoctoral Fellow
University of British Columbia

Convex Integration for Nonlinear Partial Differential Equations

Tuesday, February 18, 2014, 5:00pm
Ungar Room 402

Abstract: The so-called method of convex integration has been recognized as a powerful tool in various fields of mathematics, including geometry and, more recently, partial differential equations and the calculus of variations. I will discuss the method and two of its recent applications: On the one hand, starting with the recent work of C. De Lellis and L. Szkelyhidi, convex integration methods have been used to construct solutions of equations from fluid dynamics with surprising properties. On the other hand, together with K. Koumatos and F. Rindler, we were able to construct unexpected solutions to first-order equations involving the Jacobian determinant, motivated by questions arising in nonlinear elasticity theory.

Dusa McDuff
Helen Lyttle Kimmel '42 Professor of Mathematics
Barnard College, Columbia University

Symplectic Embeddings in Dimensions 4 and Above

Tuesday, February 11, 2014, 5:00pm
Ungar Room 402

Abstract: Symplectic geometry is a fascinating mix of flexibility and rigidity. One way in which this is shown is in the properties of symplectic embeddings. After explaining what a symplectic structure is, I will describe some recent (and not so recent) results about this question.

Dr. Ting Zhou
C.L.E. Moore Instructor

On Transformation-Optics Based Invisibility

Monday, January 27, 2014, 5:00pm
Ungar Room 402

Abstract: In this talk, I shall discuss the transformation optics based design of electromagnetic invisible cloaks from the inverse problems point of view. In order to avoid the difficulty posed by the singular structure required for ideal cloaking, we study the regularized approximate cloaking in prototypical models, for time harmonic Maxwell's equations in R3 and the scalar Helmholtz equations in R2. In particular, as the regularization parameter converges to zero, i.e., as the approximate cloaking converges to the ideal one, we will see that different types of boundary conditions appearing at the interior of the cloaking interface. Some of them is of non-local pseudo-differential type.

Mihaela Ignatova
Stanford University

On Well-posedness for Free Boundary Fluid-structure Interaction Models

Friday, January 24, 2014, 5:00pm
Ungar Room 402

Abstract: We address a fluid-structure interaction model describing the motion of an elastic body immersed in a viscous incompressible fluid. The model consists of the Navier-Stokes equations (for the fluid) and a linear system of elasticity (for the elastic solid) coupled on the free moving interface via natural dynamic and transmission boundary conditions. We first derive a priori estimates for the local existence of solutions for a class of initial data which also guaranties uniqueness, which leads to the local well-posedness of the model with less regular initial data then previously known. Moreover, under additional interior damping and stabilization terms on the free interface, we prove the global existence and exponential decay of solutions provided the initial data is sufficiently small.

Collaborators: I. Kukavica, I. Lasiecka, and A. Tuffaha

Gregory Pearlstein
Texas A&M University

Singular Metrics and the Hodge Conjecture

Tuesday, January 21, 2014, 5:00pm
Ungar Room 402

Abstract: A Hodge class on a smooth complex projective variety gives rise to an associated hermitian line bundle on a Zariski open subset of a complex projective space P^n. I will discuss recent work with P. Brosnan which shows that the Hodge conjecture is equivalent to the existence of a particular kind of degenerate behavior of this metric near the boundary.

Professor M. Teicher
Bar-Ilan University

Computational Aspects of the Braid Group and Applications

Friday, January 17, 2014, 5:00pm
Ungar Room 402

Abstract: In the talk we will present the 3 most difficult problems in the braid group – "The Word Problem", "The Conjugacy Problem", "The Hurwitz Equivalence Problem" – partial solutions, and applications to cryptography.

Dr. Vedran Sohinger
University of Pennsylvania

On Some Problems Concerning the Nonlinear Schrodinger Equation

Thursday, January 16, 2014, 5:00pm
Ungar Room 402

Abstract: In this talk, we will summarize several recent results concerning the nonlinear Schrodinger equation (NLS). The first part of the talk is dedicated to the study of the low-to-high frequency cascade which occurs as a result of the NLS evolution. In particular, one wants to look at how the frequency support of a solution evolves from the low to the high frequencies. This phenomenon can be quantitatively described as the growth in time of the high Sobolev norms of the solution. We present a method to bound this growth using the idea of an almost conservation law, which was previously used in the low regularity context in the work of Bourgain and Colliander, Keel, Staffilani, Takaoka, and Tao.

In the second part of the talk, we will study the Gross-Pitaevskii hierarchy. This is an infinite system of linear partial differential equations which occurs in the derivation of the nonlinear Schrodinger equation from the dynamics of N-body Bose systems. We will study this hierarchy on the three-dimensional torus. We will show a conditional uniqueness result for the hierarchy in a class of density matrices of regularity strictly greater than 1. Our result builds on the previous study of this problem on R^3 by Erdos, Schlein, and Yau, as well as by Klainerman and Machedon and on the study of this problem on T^2 by Kirkpatrick, Schlein, and Staffilani. Finally, we will apply randomization techniques in order to study randomized forms of the Gross-Pitaevskii hierarchy at low regularities, as was done in the setting of nonlinear dispersive equations starting with the work of Bourgain. The second part of the talk is based on joint work with Philip Gressman and Gigliola Staffilani.

Dr. King-Yeung Lam
Ohio State University

Evolution of Dispersal: A Reaction-Diffusion Approach

Tuesday, January 14, 2014, 5:00pm
Ungar Room 402

Abstract: We consider a reaction-diffusion model of two competing species for the evolution of conditional dispersal in a spatially varying but temporally constant environment. Two species are different only in their dispersal strategies, which are a combination of random dispersal and biased movement upward along the resource gradient. In the absence of biased movement or advection, A. Hastings showed that dispersal is selected against in spatially varying environments. When there is a small amount of biased movement or advection, we show the existence of a positive Evolutionarily Stable Strategy in diffusion rates, which is a form of Nash Equilibrium of the underlying population game.Our analysis of the model suggests that a balanced combination of random and biased movement might be a better habitat selection strategy for populations.

Dr. Dimitri R. Yafaev
University of Rennes, France

Convolutions and Hankel Operators

Monday, December 9, 2013, 5:00pm
Ungar Room 402

Abstract: We compare two classes of integral operators: convolutions in L2(R) and Hankel operators in L2(R+). Convolutions have integral kernels b(x - y); they can be standardly diagonalized by the Fourier transform. Hankel operators H can be realized as integral operators with kernels h(t+s). They do not admit an explicit diagonalization. Nevertheless we show that they can be quasi-diagonalized as H = L*ΣL. Here L is the Laplace transform, Σ is the operator of multiplication by a function σ (λ), λ > 0. We find a scale of spaces of test functions where L acts as an isomorphism. Then L* acts as an isomorphism in the corresponding spaces of distributions. We show that h = L* σ which yields an one-to-one correspondence between kernels and sigma-functions of Hankel operators. The sigma-function σ (λ) of a self-adjoint Hankel operator H contains substantial information about its spectral properties. Thus we show that the operators H and Σ have the same numbers of positive and negatives eigenvalues. In particular, we find necessary and sufficient conditions for sign-definiteness of Hankel operators. These results are illustrated at examples of quasi-Carleman operators generalizing the classical Carleman operator with kernel h(t) = t-1 in various directions.

Dr. Dimitri R. Yafaev
University of Rennes, France

Convolutions and Hankel Operators

Monday, December 9, 2013, 5:00pm
Ungar Room 402

Abstract: We compare two classes of integral operators: convolutions in L2(R) and Hankel operators in L2(R+). Convolutions have integral kernels b(x - y); they can be standardly diagonalized by the Fourier transform. Hankel operators H can be realized as integral operators with kernels h(t+s). They do not admit an explicit diagonalization. Nevertheless we show that they can be quasi-diagonalized as H = L*ΣL. Here L is the Laplace transform, Σ is the operator of multiplication by a function σ (λ), λ > 0. We find a scale of spaces of test functions where L acts as an isomorphism. Then L* acts as an isomorphism in the corresponding spaces of distributions. We show that h = L* σ which yields an one-to-one correspondence between kernels and sigma-functions of Hankel operators. The sigma-function σ (λ) of a self-adjoint Hankel operator H contains substantial information about its spectral properties. Thus we show that the operators H and Σ have the same numbers of positive and negatives eigenvalues. In particular, we find necessary and sufficient conditions for sign-definiteness of Hankel operators. These results are illustrated at examples of quasi-Carleman operators generalizing the classical Carleman operator with kernel h(t) = t-1 in various directions.

Dr. Erik Lundberg
Purdue University

Statistics on Hilbert's Sixteenth Problem

Tuesday, December 3, 2013, 5:00pm
Ungar Room 402

Abstract: We start with a question motivated by the fundamental theorem of algebra: How many zeros of a random polynomial are real? We discuss three Gaussian ensembles that lead to three different answers. For a polynomial in several variables, the real section of its zero set is much more complicated. Hilbert's sixteenth problem asks to study the possible arrangements of the connected components, and is especially concerned with the case of many components. I will describe a probabilistic approach to studying the topology, volume, and arrangement of the zero set (in real projective space) for a Gaussian ensemble of homogeneous polynomials. We will emphasize a model for random polynomials that is built from a basis of spherical harmonics (eigenfunctions of the spherical Laplacian). This is joint work with Antonio Lerario.

Dr. Paul Kirk
Indiana University, Bloomington

Looking at Knots and 3-manifolds from the Perspective of the Space of Representations of Their Fundamental Group

Friday, November 8, 2013, 4:00pm
Ungar Room 402

Abstract: A fruitful way to study 3-dimensional manifolds, such as the complement of a knot in a 3-sphere, is to study the topological space (algebraic variety) of conjugacy classes of representations of its fundamental group to a Lie group such as SU(2) or SL(2,C). I will describe part of the history and success of this approach, with a focus on illustrating low dimensional examples.

Dr. Gregory Pearlstein
Texas A&M University

Normal Functions and the Hodge Conjecture

Tuesday, September 24, 2013, 5:00pm
Ungar Room 402

Abstract: The theory of normal functions and the Hodge conjecture have their origin in the study of algebraic cycles by Lefschetz and Poincare. I will sketch the history of the subject and discuss some of my recent work on singularities of normal functions to the Hodge conjecture and the zero locus of a normal function to a conjectural filtration of Bloch and Beilinson.

Dr. Shuangjie Peng
Department of Mathematics
Huazhong Normal University, Wuhan, China

On Schrodinger Systems with Nonlinear or Linear Coupling

Tuesday, August 20, 2013, 11:00am
Ungar Room 402

Abstract: We will talk about how to get infinitely many positive non-radial vector solutions which are synchronized or segregated for a Schrodinger system with nonlinear coupling and radially symmetric potentials. We also discuss multiplicity of vector solutions which are segregated for a Schrodinger system with linear coupling.

Dr. Yinbin Deng
Department of Mathematics
Huazhong Normal University, Wuhan China

On the Positive Radial Solutions of a Class of Singular Semilinear Elliptic Equations

Monday, August 19, 2013, 4:00pm
Ungar Room 402

Abstract: In this talk, we are concerned with the following elliptic equation

div (A(|x|)∇u) + B(|x|)up = 0        in Rn,        (0.1)

where p > 1, n ≥ 3, A(|x|) > 0 is differentiable in Rn ¥{0} and B(|x|) is a given nonnegative Hölder continuous function in Rn ¥{0}. The asymptotic behavior at infinity and structure of separation property of positive radial solutions with different initial data for (0.1) are discussed. Moreover, the existence and separation property of infinitely many positive solutions for Hardy equation and an equation related to Caffarelli-Kohn-Nirenberg inequality are obtained respectively, as special cases.

Professor Erwan Rousseau
LATP CMI Aix-Marseille University

Complex Hyperbolicity, Differential Equations and Automorphic Forms

Wednesday, April 24, 2013, 4:00pm
Ungar Room 402

Abstract: I will explain how the study of entire curves in complex projective manifolds is related to differential equations and automorphic forms.

Dr. Mu-Tao Wang
Columbia University

A Minkowski Inequality and a Penrose Inequality

Friday, April 5, 2013, 5:00pm
Ungar Room 402

Abstract: The classical inequality of Minkowski relates the total mean curvature of a convex surface to the area of the surface. I shall discuss a newly discovered Minkowski type inequality which can be interpreted as the Penrose inequality for collapsing shells in general relativity. This is joint work with Simon Brendle and Pei-Ken Hung.

Persi Diaconis
Mary V. Sunseri Professor of Statistics and Mathematics
Stanford University

Shuffling Cards, Breaking Rocks and Hopf Algebras

Friday, March 29, 2013, 5:00pm
Ungar Room 402

NOTE: Professor Diaconis will also be delivering the McKnight-Zame Distinguished Lecture on Wednesday, March 27.

Abstract: Hopf algebras are combinatorial objects introduced by topologists to study the topology of classical groups. In joint work with Amy Pang and Arun Ram, we show that the Hopf square map (coproduct followed by product) often has a simple probabilistic interpretation. It allows us to explicitly diagonalize well-known Markov chains: the classical model of riffle shuffling cards and a rock breaking model of Kolmogorov among others. I will try to present all of this in mathematical English, explaining all the words above.

Susan Holmes
Professor of Statistics
Stanford University

Mathematical and Statistical Approaches to Heterogeneous Data:
Challenges from the Human Microbiome

Thursday, March 28, 2013, 5:00pm
Ungar Room 402

Abstract: Through new sequencing technologies, we can make a census of the bacteria living in the human gut. We also have phylogenetic information about the taxa present and clinical information about the subjects. Distance based methods allow us to create useful representations integrating all this data and make useful visualizations that allow us to discover bacterial markers for disease and follow the dynamics of the bacterial communities.

This is joint work with David Relman and his Stanford Lab.

Dr. Richard Stanley
Norman Levinson Professor of Applied Mathematics
Massachusetts Institute of Technology

Polynomial Sequences of Binomial Type

Tuesday, January 15, 2013, 5:00pm
Ungar Room 402

Mihalis Dafermos
Professor of Mathematical Physics
Cambridge University

A Scattering Theory Construction of Dynamical Black Holes

Wednesday, December 19, 2012, 4:00pm
Ungar Room 402

Dr. Joshua Greene
Boston College

Dehn Surgery and Floer Homology

Thursday, November 29, 2012, 5:00pm
Ungar Room 402

Abstract: Dehn surgery is a natural operation occurring in low-dimensional topology. It gives a method to create 3-manifolds out of knots and links, and there are many very attractive results and problems about it. One central problem is the Berge conjecture, which predicts when Dehn surgery along a knot in the three-sphere can produce a lens space. Floer homology has played a prominent role in low-dimensional topology during the past twenty-five years. It developed out of gauge theory and symplectic geometry, and one of its versions associates invariants to 3-manifolds and other objects in low-dimensional topology. I will discuss the central role that Floer homology has played in the study of Dehn surgery, and in particular what it tells us about the Berge conjecture.


Dr. Eugene Gorsky
Stony Brook University

Compactified Jacobians, q,t-Catalan Numbers and Knot Invariants

Thursday, November 1, 2012, 5:00pm
Ungar Room 402

Abstract: Campillo, Delgado and Gusein-Zade proved that the semigroup of a plane curve singularity encodes the information about the Alexander polynomial of its link. Oblomkov and Shende conjectured an extension of their result to to the HOMFLY polynomial, which uses the Hilbert schemes and compactified Jacobians of a singular curve. I will explain the combinatorics of their construction in the simplest example of torus knots, and relate it to the generalization of q,t-Catalan numbers of Garsia and Haiman. The talk is based on joint work with M. Mazin.

Dr. Fernando Schwartz
University of Tennessee

Geometric Inequalities for Hypersurfaces

Thursday, October 11, 2012, 5:00pm
Ungar Room 402

Abstract: We revisit some classic estimates for the capacity as well as a version of the Alexandrov-Fenchel inequality for hypersurfaces of Euclidean space. We provide new, more general proofs of these inequalities, and include some rigidity statements. The results are joint work with Alexandre Freire.

Professor Lars Andersson
Albert Einstein Institute
Max Planck Institute for Gravitational Physics

Cosmological Models and Stability

Thursday, October 4, 2012, 5:00pm
Ungar Room 402

Abstract: In this talk I will discuss some mathematical results on inhomogeneous cosmological models, focusing on late time behavior and the issues of nonlinear stability versus instability.

Professor Fernando Coda Marques
IMPA, Brazil

Min-max Minimal Surfaces and the Willmore Conjecture

Monday, April 30, 2012, 4:00pm
Ungar Room 402

Abstract: In 1965, T. J. Willmore conjectured that the integral of the square of the mean curvature of a torus immersed in Euclidean three-space is at least 2. In this talk we will discuss a proof of this conjecture that uses the min-max theory of minimal surfaces. This is a joint work with Andre Neves of Imperial College (UK).

Professor Ernesto Lupercio
Center for Research and Advanced Studies of the
National Polytechnic Institute (Cinvestav-IPN)

Virtual Orbifold Cohomology

Wednesday, April 25, 2012, 4:00pm
Ungar Room 402

Abstract: In this talk of a mostly expository nature I will explain the orbifolding procedure for topological field theories and a new family of examples. I will first introduce the concept of topological field theory and orbifold. This is joint with Gonzalez, Segovia, and Uribe.

Dr. Sebastian Schreiber
University of California, Davis

Should I Stay or Should I Go?
An SDE Perspective on the Ecology and Evolution of Movement

Friday, March 30, 2012, 5:00pm
Ungar Room 402

Abstract: All populations, whether they be plants, animals, or viruses, live in spatially and temporally variable environments. Understanding how this variability influences population persistence and the evolution of movement is a fundamental issue of practical and theoretical importance in population biology. Prior work (including important contributions due to Chris Cosner and Steve Cantrell) has shown that spatial variability, in and of itself, enhances persistence and selects against random movement as well as movement into sink habitats (places unable to harbor a self-sustaining population). Alternatively, temporal variability, in and of itself, inhibits persistence and exerts no selective pressures on movement. The combined effects, however, of spatial and temporal variability are remarkably complex. This combined variability can select for movement into sink habitats and allow for populations to persist in landscapes comprised solely of sink habitats. In this talk, I will discuss recent analytic results in which populations living in patchy environments are modeled using stochastic differential equations (SDEs). These results provide a diversity of new insights into population persistence and the evolution of movement. Part of this work was done in collaboration with Steve Evans (Berkeley), Peter Ralph (Davis), and Arnab Sen (Cambridge).

Dr. Ryan Derby-Talbot
Quest University

Are Complicated 3-manifolds Complicated?

Thursday, March 29, 2012, 5:00pm
Ungar Room 402

Abstract: In this talk we will discuss several ways that one can construct 3-manifolds (shapes like our spatial universe), and various ways that these constructions can be made complicated. Surprisingly, a 3-manifold made complicated by one construction may actually be simple in another construction, and necessarily so. We will explore different ways that complicating 3-manifolds in some ways simplifies them in others. Lots of fun pictures are promised.

Professor Lars Andersson
Albert Einstein Institute
Max Planck Institute for Gravitational Physics

Hidden Symmetries and Conserved Charges

Monday, March 12, 2012, 4:00pm
Ungar Room 402

Abstract: Test fields with non-zero spin, eg. Maxwell and linearized gravity provide an important model problem for black hole stability. Fields with non-zero spin admit non-radiating modes which must be eliminated in order to prove decay. In this talk I will discuss the relation between conserved charges and hidden symmetries for linearized gravity on Minkowski space and vacuum spaces of Petrov type D and outline the application of these ideas in proving estimates for the higher spin fields on the Kerr background.

Dr. Andras Stipsicz
Renyi Institute of Mathematics (Budapest)
Institute for Advanced Study (Princeton)

Computations of Heegaard Floer Homologies

Friday, March 9, 2012, 4:00pm
Ungar Room 402

Abstract: Heegaard Floer homology groups were recently introduced by Ozsvath and Szabo to study properties of 3-manifolds and knots in them. The definition of the invariants rests on delicate holomorphic geometry, making the actual computations cumbersome. In the lecture we will recall the basic definitions and theorems of the theory, and show how to define the simplest version in a purely combinatorial manner. For a special class of 3-manifolds the more general version will be presented by simple combinatorial ideas through lattice homology of Nemethi.


Professor Ross Pinsky
Technion – Israel Institute of Technology

Probabilistic and Combinatorial Aspects of the Card-Cyclic to Random Insertion Shuffle

Tuesday, March 6, 2012, 5:00pm
Ungar Room 402

Abstract: Consider a permutation σj ∈ Sn as a deck of cards numbered from 1 to n and laid out in a row, where σj denotes the number of the card that is in the j-th position from the left. We study some probabilistic and combinatorial aspects of the shuffle on Sn defined by removing and then randomly reinserting each of the n cards once, with the removal and reinsertion being performed according to the original left to right order of the cards. The novelty here in this nonstandard shuffle is that every card is removed and reinserted exactly once. The bias that remains turns out to be quite strong and possesses some surprising features.

Dr. Carmen Coll
Instituto de Matematica Multidisciplinar
Universidad Politecnica de Valencia
Valencia, Spain
ITME Visiting Scholar

Identifiability of Parameters for Structured Systems

Monday, February 27, 2012, 3:30pm
Ungar Room 402

Abstract: Mathematical models have been successfully developed to study real processes. Usually, the situation is as follows: the inputs, u, can be controlled, the outputs, y, can be observed, but the description of what happens inside is not unknown. This situation is called the input-output behavior of the system.

In general, the state-space representation of these processes involves coefficient matrices and the equations of the model have unknown parameters that can be determined from experimental data. A good formulation of the model is essential because this fact allows us to predict or control the behavior of the real process. Moreover, it can be used to estimate quantities that cannot be measured directly from observations. In this case, it is useful to obtain these parameters to accurate the model. The problem of identifying the unknown parameters within the model uniquely from the experiment considered is called the identifiability problem. That is, a system is identifiable if the relationship between the set of possible parameter values and the set of possible input-output behaviors is one-to-one.

On the other hand, many of these models have a fixed structure and we want to solve the identifiability problem when this structured holds on. The structural identifiability analysis of a system is a preliminary analysis of the model structure oriented to the parameter and model identification. The problem of the structural identifiability of the model consists of the determination of all parameter sets which give the same input-output structure.

In this talk, parametric systems with different matrix structures are considered. The structural properties of the model are studied, and some conditions to assure the structural identifiability are given. These results guarantee the existence of only one solution for the parameters of the system. In practice, systems with one of these structures arise, for example, via discretization or finite difference methods for solving boundary and initial value problems involving differential or partial differential equations.

Dr. Huaiping Zhu
York University

Forecasting Mosquito Abundance and West Nile Virus Risk Using Weather and Environment Conditions

Friday, February 24, 2012, 5:00pm
Ungar Room 402

Abstract: It has been witnessed for the last decades that climate change has great impact on the emerging and reemerging of vector-borne diseases, yet it must be admitted that the actual impact of climate change on vector population and diseases transmission are still far from clear. In this talk, I will present a modeling study of the West Nile virus in the Peel region of Ontario, Canada. By using surveillance data, weather data and land use information, we develop both statistical and dynamical models incorporating weather conditions and land use information for the vector-mosquitoes abundance and risk assessment of West Nile virus. I will discuss the statistical properties of the dynamical models and present a collaborative effort with Peel region and Public Health Agency of Canada in developing tools for forecasting the mosquito-abundance and the virus risk.

Dr. Lev Ginzburg
Department of Ecology and Evolution
Stony Brook University

Life is 4D: Allometric Slopes Are Understandable when Viewed in 4D

Monday, February 6, 2012, 3:30pm
Ungar Room 402

Professor Roger Arditi
Department of Ecology and Evolution
Université Pierre et Marie Curie, Paris

How Species Interact

Friday, February 3, 2012, 5:00pm
Ungar Room 402

Abstract: Understanding the functioning of ecosystems requires the understanding of the interactions between consumer species and their resources. How do these interactions affect the variations of population abundances? How do population abundances determine the impact of predators on their prey? The authors defend the view that the "null model" that most ecologists tend to use (derived from the Lotka-Volterra equations) is inappropriate because it assumes that the amount of prey consumed by each predator is insensitive to the number of conspecifics. The authors argue that the amount of prey available per predator (rather than the absolute abundance of prey) is the basic determinant of the dynamics of predation. This so-called ratio dependence is shown to be a much more reasonable "null model". Lessons can be drawn from a similar debate that took place in microbiology in the 1950's. Currently, populations of bacteria are known to follow the analogue of ratio dependence when growing in real-life conditions. Three kinds of arguments are developed. First, it is shown that available direct measurements of prey consumption are "in the middle" but most are close to ratio dependence and all are clearly away from the usual Lotka-Volterra relationship; an example is the system of wolves and moose on Isle Royale. Second, indirect evidence is based on the responses of food chains to nutrient enrichment: all empirical observations at the community level agree very well with the ratio-dependent view. Third, mechanistic approaches explain how ratio dependence emerges at the global scale, even when assuming Lotka-Volterra interactions at the local scale; this is illustrated by microcosm experiments, by individual-based models and by mathematical models. Changing the fundamental paradigm of the predator-prey interaction has far-reaching consequences, ranging from the logical consistency of theoretical ecology to practical questions of eco-manipulation, biological control, conservation ecology.

This work is in collaboration with Lev Ginzburg.

Professor Larry Shepp
Patrick T. Harker Professor
Wharton School, University of Pennsylvania
Board of Governor's Professor
Rutgers University

Is Mathematical Modeling Able to Give Insight into Current Questions in Finance, Economics, and Politics?

Thursday, February 2, 2012, 5:00pm
Ungar Room 402

Abstract: Part I. I argue that rigorous mathematics gives insight into the current question of whether taxation helps or hinders employment.

Part II. I compute (within one simple model) how much money future knowledge, obtained either via insider information or via high-frequency trading, of future stock prices brings to a possessor of such knowledge.

Professor Sergiu Klainerman
Higgins Professor of Mathematics
Princeton University

On the Bounded L2 Curvature Conjecture as a Breakdown Criteria for the Einstein Vacuum Equations

Wednesday, January 11, 2012, 4:00pm
Ungar Room 402

Abstract: I will talk about my recent work with Rodnianski and Szeftel concerning a solution of the conjecture. I will also compare the result with the other known breakdown criteria in GR.

Dr. Michael Eichmair
Massachusetts Institute of Technology

Isoperimetric Structure of Initial Data Sets

Friday, December 16, 2011, 4:00pm
Ungar Room 402

Abstract: I will present joint work with Jan Metzger. A basic question in mathematical relativity is how geometric properties of an asymptotically flat manifold (or initial data set) encode information about the physical properties of the space time that it is embedded in. For example, the square root of the area of the outermost minimal surface of an initial data with non-negative scalar curvature provides a lower bound for the "mass" of its associated space time, as was conjectured by Penrose and proven by Bray and Huisken-Ilmanen. Other special surfaces that have been studied in this context include stable constant mean curvature surfaces and isoperimetric surfaces. I will explain why positive mass works to the effect that large stable constant mean curvature surfaces are always isoperimetric. This answers an old conjecture of Bray's and complements the results by Huisken-Yau and Qing-Tian on the "global uniqueness problem for stable CMC surfaces" in initial data sets with positive scalar curvature. Time permitting, I will sketch applications related to G. Huisken's isoperimetric mass and very recent related results with S. Brendle on further isoperimetric features of the exact spatial Schwarzschild metric.

Dr. Martin Bootsma
Department of Mathematics
Utrecht University, The Netherlands

The Spreading Capacity of Methicillin-resistant Staphylococcus aureus (MRSA)

Tuesday, December 6, 2011, 5:00pm
Ungar Room 402

Abstract: In my talk I will discuss an example of cross-fertilization between medicine and mathematics. Data on the size of outbreaks with methicillin-resistant Staphylococcus aureus (MRSA) in Dutch hospitals were collected to estimate the transmissibility of the two most relevant MRSA strains in the Netherlands. Analysis of these data led to a relation between the epidemiological model for the spread of MRSA in hospitals and queuing theory, and a new estimation method for transmissibility of pathogens in outbreak settings with contact screening. In very recent theoretical work, Amaury Lambert and Pieter Trapman derive an improved estimator if easily collectable data is available. These data are now collected in a currently performed study in the Netherlands.

Dr. Pierre Magal
University of Bordeaux Segalen, France

P-gp Transfer and Acquired Multi-drug Resistance in Tumors Cells

Monday, November 28, 2011, 4:30pm
Ungar Room 402

Abstract: Multi-drug resistance for cancer cells as been a serious issue since several decades. In the past, many models have been proposed to describe this problem. These models use a discrete structured for the cancer cell population, and they may include some class of resistant, non resistant, and acquired resistant cells. Recently, this problem has received a more detailed biological description, and it turns out that the resistance to treatments is due in 40% of cancers to a protein called P-glycoprotein (P-gp). Moreover some new biological experiments show that transfers can occur by the mean of Tunneling nanoTubes built in between cells (direct transfers). Transfers can also occur through microparticles (containing P-pg) released by over expressing cells into the liquid surrounding these cells. These microparticles can then diffuse and can be recaptured by the cells (indirect transfers). This transfers turn to be responsible for the acquired resistance of sensitive cells. The goal of this talk is to introduce this problem, and to present a cell population dynamic model with continuous P-gp structure.

Professor DaGang Yang
Tulane University

Einstein 4-Manifolds with Pinched Sectional Curvature

Thursday, November 17, 2011, 5:00pm
Ungar Room 402

Abstract: Let (M, g) be a compact, simply connected n-dimensional Riemannian manifold with sectional curvature K. (M, g) is said to be pointwise ε-pinched for some constant 1 ≥ ε > 0 if there is a positive function K0 on M such that K0 ≥ K > εK0.

Question: For what values of ε, can one expect the Ricci flow initiated from g to converge to a metric of constant sectional curvature, and therefore diffeomorphic to the standard sphere Sn?

The 1/4-pinched differentiable sphere theorem, by H.W. Chen for n = 4, and by S. Brendle and R. M. Schoen for n ≥ 5, says that ε = 1/4 is the smallest possible value. For ε < 1/4, P. Petersen and T. Tao have shown that there is a constant εn, 1/4 > εn > 0, for each dimension n such that, if 1 ≥ K > εn, then the Ricci flow initiated from g will converge to a metric which is either of constant sectional curvature or is a compact rank one symmetric space.

It is therefore natural to propose the following question: For each n ≥ 4, what is the smallest pinching constant εn > 0 such that, if K0 ≥ K > εnK0, then the Ricci flow initiated from g can still be expected to converge to a metric of constant sectional curvature or to a metric of a compact rank one symmetric space?

In other words, are there any other models of Einstein manifolds with pinched positive sectional curvature in each dimension n ≥ 4?

In this talk, I shall discuss some old and new results in this area for n=4.

Dr. Jan Medlock
Clemson University

Issues in the Ecology and Evolution of Dengue

Friday, April 8, 2011, 5:00pm
Ungar Room 402

Abstract: Dengue is a mosquito-borne viral pathogen that causes large amounts of disease in the tropics and sub-tropics. Dengue viruses are divided into four large clades, called serotypes: infection with a virus produces complete immunity to viruses within that same serotype, but increases the risk of severe disease upon infection with a virus from a different serotype. Multiple mechanisms have been hypothesized for this interaction between serotypes in the human immune system, which, combined with seasonal oscillations in mosquito abundances, lead to complex behavior in mathematical models. In addition, two new interventions for dengue are currently in intense development: a vaccine that protects against all four serotypes and transgenic mosquitoes that are less-suitable vectors. In this talk, I will discuss a model for evolution of dengue viruses in response to these new interventions and work in progress on the best population groups to target with vaccine to minimize disease burden.

Dr. Jim Haglund
University of Pennsylvania

Macdonald Polynomials and the Hilbert Series of the Quotient Ring of Diagonal Coinvariants

Thursday, April 7, 2011, 5:00pm
Ungar Room 402

Abstract: Macdonald polynomials are symmetric functions in a set of variables X which also depend on two parameters q,t. In this talk we describe how a formula of Haiman for the Hilbert series of the quotient ring of diagonal coinvariants in terms of Macdonald polynomials implies a much simpler expression for the Hilbert series involving matrices satisfying certain constraints.

Dr. Marcus Khuri
Stony Brook University

The Positive Mass Theorem with Charge Revisited

Thursday, March 31, 2011, 5:00pm
Ungar Room 402

Abstract: In the early 80's Hawking et al. generalized the positive mass theorem to include charge. It was conjectured that the case of equality should occur only for the extremal black hole solutions known as Majumdar-Papapetrou spacetimes. Chrusciel et al. confirmed this under extra assumptions. In this talk we will show how these extra hypotheses may be removed. This is joint work with Gilbert Weinstein.

Richard P. Stanley
Norman Levinson Professor of Applied Mathematics
Massachusetts Institute of Technology

A Survey of Alternating Permutations

Tuesday, March 22, 2011, 5:00pm
Ungar Room 402

Professor Xiaodong Wang
Michigan State University

Volume Entropy and Ricci Curvature

Thursday, March 10, 2011, 5:00pm
Ungar Room 402

Abstract: The volume entropy is a very interesting invariant of a Riemannian manifold. When the Ricci curvature has a negative lower bound, there is a sharp lower bound for the volume entropy. I will discuss why the equality case characterizes hyperbolic manifolds. In certain cases, we can also prove that the manifold is close to a hyperbolic manifold in the Gromov-Hausdorff sense if the volume entropy is close to the sharp lower bound. The method involves the Busemann compactification and Patterson-Sullivan measure. This is a joint work with Francois Ledrappier.

Dr. David Smith
University of Florida
Center for Disease Dynamics Economics and Policy Washington, DC

Recasting the Theory of Transmission by Mosquitoes

Thursday, February 24, 2011, 5:00pm
Ungar Room 402

Abstract: Mathematical modeling for mosquito-borne diseases has been used to develop theory and guide disease control for more than a century, but the demands on models have been changing. Analysis of a comprehensive review of mosquito-borne transmission models demonstrated that mosquito-borne disease models follow the conventions of the Ross-Macdonald model and that there has been little innovation in modeling transmission. A new mathematical description of transmission was based on mosquito movement, aquatic ecology and blood feeding behavior. Mosquito movement can be described concisely as a random walk on a bipartite graph. Transmission also depends on the ways that mosquitos allocate bites on humans and the way humans allocate their time at risk. This framework provides a starting point for reformulating a new theory of transmission that captures other aspects of mosquito behavior that are important for transmission but absent from the Ross-Macdonald model.

Dr. Xinzhi Liu
Department of Applied Mathematics
University of Waterloo
Waterloo, Canada

Epidemic Models with Switching Parameters

Wednesday, February 23, 2011, 5:00pm
Ungar Room 402

Abstract: Epidemic models are vital for implementing, evaluating, and optimizing control schemes in order to eradicate a disease. These mathematical models may be oversimplified, but they are useful for gaining knowledge of the underlying mechanics driving the spread of a disease, and for estimating the number of vaccinations required to eradicate a disease. This talk discusses some epidemic models with switching parameters. Both constant control and pulse control schemes are examined, and, in doing so, we hope to gain insight into the effects of a time-varying contact rate on critical control levels required for eradication.

Professor Tatiana Toro
University of Washington

Potential Theory Meets Geometric Measure Theory

Tuesday, February 22, 2011, 5:00pm
Ungar Room 402

Abstract: A central question in Potential Theory is the extent to which the geometry of a domain influences the boundary regularity of solutions to divergence form elliptic operators. To answer this question one studies the properties of the corresponding elliptic measure. On the other hand one of the central questions in Geometric Measure Theory (GMT) is the extent to which the regularity of a measure determines the geometry of its support. The goal of this talk is to present a few instances in which techniques from GMT and Harmonic Analysis come together to produce new results in both of these areas.

Professor Ernesto Lupercio
Research and Advanced Studies Center of the
National Polytechnic Institute of Mexico (Cinvestav-IPN)
Winner of the 2009 Srinivasa Ramanujan Prize

The Moduli Space of (Non-commutative) Toric Varieties

Friday, February 18, 2011, 4:30pm
Ungar Room 402

Abstract: In this talk I will describe my work in progress with Laurent Meersseman and Alberto Verjovvsky on the moduli space of Toric Manifolds. Using specific families of foliations and the Gale transform we describe some basic geometric and topological properties of this moduli space.

Professor Herbert S. Wilf
Thomas A. Scott Emeritus Professor of Mathematics
University of Pennsylvania

There's Plenty of Time for Evolution

Thursday, February 17, 2011, 2:00pm
Ungar Room 402

Abstract: Those who are skeptical of the Darwinian view of evolution often argue that since there are K^n possible n letter words over a K letter alphabet, it must take an exponentially long time before random mutations of the letters will produce "the right word." We show that if the effects of natural selection are taken into account in a reasonable way, the K^n time estimate can be replaced by Kn log n. As a byproduct we obtain the mean of the largest of many geometrically distributed random variables. This is joint work with Warren Ewens.

Dr. Nathan Geer
Utah State University

The Colored Jones Polynomial and Some of Its Relatives

Tuesday, February 15, 2011, 5:00pm
Ungar Room 402

Abstract: In this talk I will discuss a problem in quantum topology called the Volume Conjecture. This conjecture relates the colored Jones polynomials with the hyperbolic volume of the knot complement. As I will explain the Volume Conjecture links together elements of topology, geometry and algebra. I will begin with a gentle introduction to knot theory and the definition of the Jones polynomial. Then I will show how to compute the colored Jones polynomial using algebra. Finally after stating the conjecture, I will discuss some related topological invariants.

Professor Larry Shepp
Wharton School, University of Pennsylvania
Rutgers University

Some New Probability Problems; Some New Solutions

Thursday, February 10, 2011, 5:00pm
Ungar Room 402

Abstract: I will first update the situation discussed last year re the artificial pancreas project. The problem is now to upgrade the sensor by getting a better algorithm for recalibration. This is a really important and beautiful problem. I will tell you about my recent work on it; it's still very open.

The second update is a beautiful theoretical problem recently posed by Mike Steele. The problem is this: consider a sequence of n iid uniform variables on [0,1]. Call a subsequence "upsy-downsy", if no three successive terms of the subsequence are monotonic. It is known (Houdre-Stanley-Widom) that if one can search all subsequences, then the expected length of the longest upsy-downsy subsequence is asymptotic to n times 2/3. Steele asked a question of interest to people in stochastic optimization: what is the length of the upsy-downsy subsequence if each term is optimally chosen, {\em without knowing the rest of the sequence, as in the famous secretary problem. We showed the answer is asymptotic to n times c, where c = 2 - \sqrt{2} = .586\ldots < 2/3$. I will indicate two approaches to this problem, each of which gives the right answer, but only one of which I regard as mathematically legitimate.

Dr. Slawomir Kwasik
Tulane University

Souls of Manifolds via Curvature and Surgery

Thursday, January 27, 2011, 5:00pm
Ungar Room 402

Abstract: Deep connections between topology and geometry will be discussed in the case of manifolds with non-negative (sectional) curvature. Historical perspective of these connections and new developments will be presented.

Dr. Chuan Xue
Mathematical Biosciences Institute at Ohio State

A Mathematical Model of Chronic Wounds

Thursday, January 20, 2011, 5:00pm
Ungar Room 402

Abstract: Chronic wound healing is a staggering public health problem, affecting 6.5 million individuals annually in the U.S. Ischemia, caused primarily by peripheral artery diseases, represents a major complicating factor in the healing process. In this talk, I will present a mathematical model of chronic wounds that represents the wounded tissue as a quasi-stationary Maxwell material, and incorporates the major biological processes involved in the wound closure. The model was formulated in terms of a system of partial differential equations with the surface of the open wound as a free boundary. Simulations of the model demonstrate how oxygen deficiency caused by ischemia limit macrophage recruitment to the wound-site and impair wound closure. The results are in tight agreement with recent experimental findings in a porcine model. I will also show analytical results of the model on the large-time asymptotic behavior of the free boundary under different ischemic conditions of the wound.

Dr. Valerie Hower
University of California, Berkeley

A Shape-based Method for Determining Protein Binding Sites in a Genome

Tuesday, January 18, 2011, 5:00pm
Ungar Room 402

Abstract: We present a new algorithm for the identification of bound regions from ChIP-Seq experiments. ChIP-Seq is a relatively new assay for measuring the interactions of proteins with DNA. The binding sites for a given protein in a genome are "peaks" in the data, which is given by an integer-valued height function defined on the genome. Our method for identifying statistically significant peaks is inspired by the notion of persistence in topological data analysis and provides a non-parametric approach that is robust to noise in experiments. Specifically, our method reduces the peak calling problem to the study of tree-based statistics derived from the data. The software T-PIC (Tree shape Peak Identification for ChIP-Seq) is available at and provides a fast and accurate solution for ChIP-Seq peak finding.

Dr. Peter Kim
University of Utah

Imatinib Dynamics and Cancer Vaccines:
From Agent-Based Models to PDEs

Thursday, January 13, 2011, 5:00pm
Ungar Room 402

Abstract: Various models exist for the interaction between the drug imatinib and chronic myelogenous leukemia. However, the role of the immune response during imatinib treatment remains unclear. Based on experimental data, we hypothesize that imatinib gives rise to a brief anti-leukemia immune response as patients enter remission.

We propose that cancer vaccinations during imatinib treatment can boost the existing immune response and lead to a sustained remission or a potential cure. To examine this hypothesis, we take a model by Michor et al. and extend it to a delay differential equation (DDE) model by incorporating an anti-leukemia immune response. We show that properly-timed vaccines can sustain the immune response to potentially prolong remission or eliminate cancer.

For comparison, we analyze an agent-based model developed independently by Roeder et al. We develop a partial differential equation (PDE) model that captures the same behavior as the Roeder agent-based model and extend it by incorporating an immune response. We conclude that both the DDE and PDE models exhibit similar behaviors with regard to cancer remission, implying that anti-leukemia immune responses may play a role in leukemia treatment.

Professor Nicolai Reshetikhin
University of California, Berkeley

Understanding Random Surfaces

Monday, December 20, 2010, 4:30pm
Ungar Room 402

Abstract: There is a bijection between a class of piece-wise linear surfaces and dimer configurations on planar graphs. A dimer configuration on a graph is a perfect matching on vertices connected by edges. Dimers are well known in biology, chemistry and statistical mechanics. For certain very natural probability measures on dimer configurations, important correlation functions can be computed as Pfaffians of N\times N matrices. This reduces the statistics of such special random surfaces to a reasonable problem in linear algebra. This allows to study such random surfaces corresponding tolarge graphs. The talk will outline this story and at the end the discussion will focus on the "continuum limit" of such random surfaces.

Professor Pierre Magal
University of Bordeaux, France

Bifurcation Problems for Structured Population Dynamics Models

Monday, December 6, 2010, 4:30pm
Ungar Room 402

Abstract: This presentation is devoted to bifurcation problems for some classes of PDE arising in the context of population dynamics. The main difficulty in such a context is to understand the dynamical properties of a PDE with non-linear and non-local boundary conditions. A typical class of examples is the so called age structured models. Age structured models have been well understood in terms of existence, uniqueness, and stability of equilibria since the 80's. Nevertheless, up to recently, the bifurcation properties of the semiflow generated by such a system has been only poorly understood.

In this presentation, we will start with some results about existence and smoothness of the center mainfold, and we will present some general Hopf bifurcation results applying to age structured models. Then we will turn to normal theory in such a context. The point here is to obtain formula to compute the first order terms of the Taylor expansion of the reduced system.

Dr. Steven White
Centre for Ecology & Hydrology
Wallingford, Oxon, UK

Controlling Mosquitoes by Classical or Transgenic Sterile Insect Techniques

Monday, November 22, 2010, 4:00pm
Ungar Room 402

Abstract: For centuries, humans have attempted to control insect populations. This is in part because of the significant mortality and morbidity burden associated with insect vector-borne diseases, but also due to the huge economic impact of insect pests leading to losses in global food production. The development of transgenic technologies, coupled with sterile insect techniques (SIT), is being explored in relation to new approaches for the biological control of insect pests.

In this talk, I explore the impact of two control strategies (classical SIT and transgenic late-acting bisex lethality) using a stage-structured mathematical model, which is parameterized for the mosquito Aedes aegypti, which can spread yellow fever, dengue fever and Chikungunya disease. Counter to the majority of studies, I use realistic pulsed release strategies and incorporate a fitness cost, which is manifested as a reduction in male mating competitiveness.

I will explore the timing of control release in constant and cyclic wild-type mosquito populations, and demonstrate that this timing is critical for effective pest management. Furthermore, I will incorporate these control strategies into an integrated pest management program (IPM) and find the optimal release strategy. Finally, I will extend the models to a spatial context, determining conditions for the prevention of mosquito invasion by the use of a barrier wall.

Dr. Kate Petersen
Florida State University

The Euclidean Algorithm and Primitive Roots

Thursday, November 18, 2010, 5:00pm
Ungar Room 402

Abstract: Artin's famous primitive root conjecture states that if n is an integer other than -1 or a square, then there are infinitely many primes p such that n is a primitive root modulo p. Although this conjecture is not known to hold for any value of n, Hooley proved it to be true under the assumption of the generalized Riemann hypothesis (GRH). We will discuss a number field version of this conjecture and its connection to the following Euclidean algorithm problem. Let O be the ring of integers of a number field K. It is well-known that if O is a Euclidean domain, then O is a unique factorization domain. With the exception of the imaginary quadratic number fields, it is conjectured that the reverse implication is true. This was proven by Weinberger under the assumption of the GRH. We will discuss recent progress towards the unconditional resolution of the Euclidean algorithm problem and the related primitive root problem. This is joint work with M. Ram Murty.

Dr. Shiwang Ma
Nankai University, China

Bounded and Unbounded Motions for Asymmetric Oscillators at Resonance

Monday, November 8, 2010, 4:30pm
Ungar Room 402

Abstract: In this talk, we consider the boundedness and unboundedness of solutions for the asymmetric oscillator

x" + ax+ - bx- + g(x) = p(t),

where x+ = max{x,0},x- = max{-x,0}, a and b are two positive constants, p(t) is a 2π-periodic smooth function and g(x) satisfies lim|x|→+∞x-1g(x) = 0. We establish some sharp sufficient conditions concerning the boundedness of all the solutions and the existence of unbounded solutions. Unlike many existing results in the literature where the function g(x) is required to be a bounded function with asymptotic limits, here we allow g(x) be unbounded or oscillatory without asymptotic limits. Some critical cases will also be considered.

Dr. Igor Rodnianski
Princeton University

Evolution Problem in General Relativity

Wednesday, November 3, 2010, 5:00pm
Ungar Room 402

Abstract: The talk will introduce basic mathematical concepts of General Relativity and review the progress, main challenges and open problems, viewed through the prism of the evolution problem. I will illustrate interaction of Geometry and PDE methods in the context of General Relativity on examples ranging from incompleteness theorems and formation of trapped surfaces to stability problems.

Dr. Lars Andersson
Albert Einstein Institute

The Black Hole Stability Problem

Friday, October 22, 2010, 4:00pm
Ungar Room 402

Abstract: The problem of nonlinear stability for the Kerr model of a rotating black hole is one of the central problems in general relativity. The analysis of linear fields on the Kerr spacetime is an important model problem for full nonlinear stability. In this talk, I will present recent work with Pieter Blue which makes use of the hidden symmetry related to the Carter constant to circumvent these difficulties and give a "physical space" approach to estimates for the wave equation, including energy bounds, trapping, and dispersive estimates. I will also discuss the field equations for higher spin fields including linearized gravity.

Dr. Yuan Lou
Ohio State University

Persistence of a Single Phytoplankton Species

Thursday, October 21, 2010, 4:30pm
Ungar Room 402

Abstract: Phytoplankton need light to grow. However, most of phytoplankton are heavier than water, so they sink. How can phytoplankton persist? We investigate a nonlocal reaction-diffusion-advection equation which models the growth of a single phytoplankton species in a water column where the species depends solely on light for its metabolism. We study the effect of sinking rate, water column depth and vertical turbulent diffusion rate on the persistence of a single phytoplankton species. This is based upon a joint work with Sze-Bi Hsu, National Tsing-Hua University.

Professor Nick Loehr
Virginia Tech

Macdonald Polynomials in Representation Theory and Combinatorics

Friday, October 8, 2010, 4:30pm
Ungar Room 402

Abstract: This talk surveys some recent work in algebraic combinatorics that illustrates surprising connections between representation theory and enumerative combinatorics. We describe how to calculate the Hilbert series of various spaces of polynomials (harmonics, diagonal harmonics, and Garsia-Haiman modules) using combinatorial statistics on permutations and parking functions. This leads to a discussion of the algebraic and combinatorial significance of the Macdonald polynomials, which have played a central role in the theory of symmetric functions for the past two decades.

Brian J. Coburn, Ph.D.
Center for Biomedical Modeling
Semel Institute of Neuroscience and Human Behavior
David Geffen School of Medicine
University of California, Los Angeles

Modeling Approaches for Influenza and HIV

Wednesday, September 8, 2010, 4:00pm
Ungar Room 402

Abstract: In this talk, I will present a survey of research projects on different mathematical models for influenza and HIV. For influenza, I will discuss two different modeling approaches. In the first approach, I will present a multi-strain/multi-host (MSMH) model that tracks the spread of inter-species strains between birds, pigs and humans. In the MSMH model, pigs are "mixing vessels" between avian and human strains and are capable of producing super-strains as a consequence of genetic recombination of these strains. I will show how specific subtypes can cause an epidemic then virtually disappear for years or even decades before reemerging (e.g., the case of H1N1). In the second approach, I will present a model that tracks the spread of influenza within flight transmission. A plane flight is much shorter scale than influenza's infectious duration; hence, we use methods from microbial risk management to assess the number of potential infections. We show that the flight duration along with the compartment will ultimately determine the passenger's risk. For HIV, I will present cross-sectional data on HIV prevalence in Lesotho, a small sub-Saharan African nation with HIV prevalence at approximately 23%. I will present our current progress on data analysis from the Health and Demographic Survey (DHS) to develop risk maps by district based on prevalence and treatment, feasibility analysis of a clinical trial, and efficacy of male circumcision as prevention for HIV.

Dr. Andrew Noble
University of Maryland

A Non-neutral Theory of Dispersal-limited Community Dynamics

Thursday, April 22, 2010, 4:30pm
Ungar Room 402

Abstract: We introduce the first analytical model of a dispersal-limited, niche-structured community to yield Hubbell's neutral theory in the limit of functional equivalence among all species. Dynamics of the multivariate species abundance distribution (SAD) for an asymmetric local community are modeled explicitly as a dispersal-limited sampling of the surrounding metacommunity. Coexistence may arise either from approximate functional equivalence or a competition-colonization tradeoff. At equilibrium, these symmetric and asymmetric mechanisms both generate unimodal SADs. Multiple modes only arise in asymmetric communities and provide a strong indication of non-neutral dynamics. Although these stationary distributions must be calculated numerically in the general theory, we derive the first analytical sampling distribution for a nearly neutral community where symmetry is broken by a single species distinct in ecological fitness and dispersal ability. Novel asymptotic expansions of hypergeometric functions are developed to make evaluations of the sampling distribution tractable for large communities. In this regime, population fluctuations become negligible. A calculation of the macroscopic limits for the symmetric and asymmetric theories yields a new class of deterministic competition models for communities of fixed-size where rescue effects facilitate coexistence. For nearly neutral communities where the asymmetric species experiences linear density-dependence in ecological fitness, strong Allee-type effects emerge from a saddle-node bifurcation at a critical point in dispersal limitation. The bistable dynamics governing a canonical Allee effect are modified by a constant influx of migrants, which raises the lower stable fixed point above zero. In the stochastic theory, a saddle-node bifurcation corresponds to the development of bimodal stationary distributions and the emergence of inflection points in plots of mean first-time to extirpation as a function of abundance.

Dr. Hao Wang
University of Alberta

The Role of Light and Nutrients in Aquatic Trophic Interactions

Friday, April 16, 2010, 4:30pm
Ungar Room 402

Abstract: Carbon (C), nitrogen (N), and phosphorus (P) are vital constituents in biomass: C supplies energy to cells, N is essential to build proteins, and P is an essential component of nucleic acids. The scarcity of any of these elements can severely restrict organism and population growth. Thus in nutrient deficient environments, the consideration of nutrient cycling, or stoichiometry, may be essential for population models. To show this idea, I will present two case studies in this talk.

We carried out a microcosm experiment evaluating competition of an invasive species Daphnia lumholtzi with a widespread native species, Daphnia pulex. We applied two light treatments to these two different microcosms and found strong context-dependent competitive exclusion in both treatments. To better understand these results we developed and tested a mechanistically formulated stoichiometric model. This model exhibits chaotic coexistence of the competing species of Daphnia. The rich dynamics of this model as well as the experiment allow us to suggest some plausible strategies to control the invasive species D. lumholtzi.

We modeled bacteria-algae interactions in the epilimnion with the explicit consideration of carbon (energy) and phosphorus (nutrient). We hypothesized that there are three dynamical scenarios determined by the basic reproductive numbers of bacteria and algae. Effects of key environmental conditions were examined through these scenarios. Competition of bacterial strains were modeled to examine Nishimura's hypothesis that in severely P-limited environments such as Lake Biwa, P-limitation exerts more severe constraints on the growth of bacterial groups with higher nucleic acid contents, which allows low nucleic acid bacteria to be competitive.

Dr. Sanja Zivanovic
Centrum Wiskunde en Informatica (CWI), Amsterdam, Netherlands

Numerical Solutions to Noisy Systems

Thursday, April 15, 2010, 5:00pm
Ungar Room 402

Abstract: We study input-affine systems where input represents some bounded noise. The system can be rewritten as differential inclusion describing the evolution. Differential inclusions are a generalization of differential equations with multivalued right-hand side. They have applications in many areas of science, such as mechanics, electrical engineering, the theory of automatic control, economical, biological, and social macrosystems. A numerical method for rigorous over-approximation of a solution set of input-affine system will be presented. The method gives high order error for a single time step and a uniform bound on the error over the finite time interval. The approach is based on the approximations of inputs by piecewise linear functions.

Dr. Priyanga Amarasekare
University of California Los Angeles

Ecological and Evolutionary Consequences of Dispersal in Multi-trophic Communities

Friday, March 12, 2010, 4:30pm
Ungar Room 402

Abstract: I investigate the effects of non-random dispersal strategies on coexistence and species distributions in multi-trophic communities with competition and predation. I conduct a comparative analysis of dispersal strategies with random and fitness-dependent dispersal at the extremes and two intermediate strategies that rely on cues (density and habitat quality) that serve as proxies for fitness. The most important finding is an asymmetry between consumer species in their dispersal effects. The dispersal strategy of inferior resource competitors that are less susceptible to predation have a large effect on both coexistence and species distributions, but the dispersal strategy of the superior resource competitor that is more susceptible to predation has little or no effect on dispersal. I explore the consequences of this asymmetry for the evolution of dispersal.

Dr. Herbert Wilf
Thomas A. Scott Emeritus Professor of Mathematics
University of Pennsylvania

How to Lose as Little as Possible

Thursday, March 11, 2010, 5:00pm
Ungar Room 402

Abstract: Suppose Alice has a coin with heads probability q and Bob has one with heads probability p > q. Now each of them will toss their coin n times, and Alice will win iff she gets more heads than Bob does. Of course, the game favors Bob, but for the given p, q, what is the choice, N(q,p), of n that maximizes Alice's chances of winning? The analysis uses the multivariate form of Zeilberger's algorithm, so a portion of the talk will be a review of the ideas underlying symbolic summation.

Dr. Mario Milman
Florida Atlantic University

Sobolev Inequalities on Probability Metric Spaces

Friday, March 5, 2010, 5:00pm
Ungar Room 402

Abstract: To formulate new Sobolev inequalities one needsto answer questions like: what is the role of dimension? What norms are appropriate to measure the integrabilitygains? Just to name a few...For example, in contrast to the Euclidean case, the integrability gains in Gaussian measure are logarithmic but dimension free (log Sobolev inequalities). So it is easy to understand the difficulties to derive a general theory. I will discuss some new methods to prove general Sobolev inequalities that unify the Euclidean and the Gaussian cases, as well as several important model manifolds.


Dr. Larry Shepp
Rutgers University
Member, National Academy of Science (NAS)
Member, Institute of Medicine (IOM)

Problems in Probability

Friday, February 26, 2010, 5:00pm
Ungar Room 402

Abstract: Several problems will be discussed: 1) What is the distribution of the empirical correlation coefficient of two (actually independent) Wiener processes? It is far from zero - correlation is induced by the arc sine law property of the sample paths. This is used by (bad) statisticians to show correlation between time series when none exists. It is a non-trivial calculation to find the actual distribution. 2) What is the relationship between the coefficients of a polynomial of degree n and the number of its real zeros? Descartes had something to say about it, but Mark Kac showed that probability theory can add a lot of insight. 3) An update on the situation discussed last year re the artificial pancreas project.

Dr. Richard Schoen
Bass Professor of Humanities and Sciences
Stanford University

Riemannian Manifolds of Constant Scalar Curvature

Tuesday, February 23, 2010, 5:00pm
Ungar Room 402

Abstract: The problem of constructing Riemannian metrics of constant scalar curvature is called the Yamabe problem. It is an important variational problem in conformal geometry, and also relates directly to the Einstein equations of general relativity. We will give a brief history and introduction to this problem and describe some new phenomena which have been discovered recently concerning issues of singular behavior and blow up of such metrics.

Dr. Pengzi Miao
School of Mathematical Sciences
Monash University

Critical Metrics for the Volume Functional on Compact Manifolds with Boundary

Thursday, February 4, 2010, 5:00pm
Ungar Room 402

Abstract: It is known that, on closed manifolds, Einstein metrics of negative scalar curvature are critical points of the usual volume functional constrained to the space of metrics of constant scalar curvature. In this talk, I will discuss how this variational characterization of Einstein metrics can be localized to compact manifolds with boundary. I will derive the critical point equation and focus on geometric properties of its solutions. In particular, if a solution has zero scalar curvature and the boundary of the manifold can be isometrically embedded into the Euclidean space as a convex hypersurface, I will show that the volume of such a critical metric is always greater than or equal to the Euclidean volume enclosed by the image of the isometric embedding, and two volumes are the same if and only if the critical metric is isometric to the Euclidean metric on a round ball. I will also give a classification of all conformally flat critical metrics. This is joint work with Luen-Fai Tam.

Dr. Brian J. Weber
RTG/Simons Center Postdoc
Stony Brook University

Einstein Metrics, the Bach Tensor, and Metric Degenerations

Monday, February 1, 2010, 5:00pm
Ungar Room 402

Abstract: One might search for "canonical metrics," such as Einstein metrics, on a manifold by trying to prove the convergence of a sequence of metrics that minimize some functional, although such a direct approach usually fails. In this talk we present an indirect approach which has been successful in some cases. A local obstruction to finding an Einstein metric in a conformal class is the non-vanishing of the Bach tensor, defined to be the gradient of the Weyl curvature functional $\int |W|^2$. On a Kaehler manifold there are no other obstructions, and any Bach-flat Kaehler metric is locally conformally Einsteinian. Additionally, the conformal factor is geometrically interesting and sometimes controllable. This talk will describe the results of a 2008 paper with X. Chen and C. LeBrun, where circumstances under which a Kaehler manifold is Bach-flat were established, and where it was shown that these conditions hold for a certain Kaehler metric on $CP^2 # 2\overline CP^2$ with non-zero conformal factor, establishing for the first time an Einstein metric on $CP^2 # 2\overline 2CP^2$.

Dr. Hans Boden
McMaster University

Metabelian SL(n,C) Representations of Knot Groups

Thursday, January 28, 2010, 5:00pm
Ungar Room 402

Abstract: In this talk, which represents joint work with Stefan Friedl, we will present a classification of irreducible metabelian SL(n,C) representations of knots groups. Under a mild hypothesis, we prove that such representations factor through a finite group, hence they are all conjugate to unitary representations, and we give a simple formula for the number of conjugacy classes. For knots with nontrivial Alexander polynomial, we discuss an existence result for irreducible metabelian representations. Given a knot group, its SL(n,C) character variety admits a natural action by the cyclic group of order n, and we show how to identify the fixed points of this action with characters of metabelian representations. If time permits, we will describe conditions under which such points are simple points in the character variety using a deformation argument of Abdelghani, Heusener, and Jebali.

Dr. Brett L. Kotschwar
C.L.E. Moore Instructor
Massachusetts Institute of Technology

Backwards Uniqueness for the Ricci Flow and the Non-expansion of the Isometry Group

Monday, January 25, 2010, 5:00pm
Ungar Room 402

Abstract: One of the fundamental properties of the Ricci flow -- an evolution equation for Riemannian metrics -- is that of isometry preservation, namely, that an isometry of the initial metric remains an isometry of the solution, at least as long as the curvature remains bounded. In this talk, I will take up the complementary problem of isometry development under the flow. While the solution may acquire new isometries in the limit, one does not expect the flow to sponsor their generation within the lifetime of the solution. The impossibility of such a phenomenon is equivalent to a backwards uniqueness (or unique-continuation) property for the equation: two solutions which agree at some non-initial time must agree identically at all previous times. I will discuss recent work which establishes this property for complete solutions of bounded curvature, and prohibits, additionally, a solution from becoming Einstein or self-similar in finite time.

Dr. Stephen Gourley
University of Surrey, UK

Impulsive Delay Equation Models for the Control of Vector-borne Diseases

Friday, January 15, 2010, 4:00pm
Ungar Room 402

Abstract: Delay equation models for the control of a vector-borne disease such as West Nile virus will be presented. The models make it possible to compare the effectiveness of larvicides and adulticides in controlling mosquito populations. The models take the form of autonomous delay differential equations with impulses (if the adult insects are culled) or a system of nonautonomous delay differential equations where the time-varying coefficients are determined by the culling times and rates (in the case where the insect larvae are culled). Sufficient conditions can be derived which ensure eradication of the disease. Eradication of vector-borne diseases is possible by culling the vector at either the immature or the mature phase. Very infrequent culling can actually lead to the mean insect population being increased rather than decreased.

Professor Philippe LeFloch
University of Paris 6 and CNRS

Einstein Spacetimes with Bounded Curvature

Thursday, December 10, 2009, 4:30pm
Ungar Room 402

Abstract: I will present recent results on Einstein spacetimes of general relativity when the curvature is solely assumed to be bounded and no assumption on its derivatives is made. One such result, in a joint work with B.-L. Chen, concerns the optimal regularity of pointed spacetimes in which, by definition, an "observer" has been specified. Under geometric bounds on the curvature and injectivity radius near the observer, there exist a CMC (constant mean curvature) foliation as well as CMC--harmonic coordinates, which are defined in geodesic balls with definite size depending only on the assumed bounds, so that the components of the Lorentzian metric has optimal regularity in these coordinates. The proof combines geometric estimates (Jacobi field, comparison theorems) and quantitative estimates for nonlinear elliptic equations with low regularity.


Dr. Zhilan Feng
Purdue University

Evolutionary Implications of Influenza Medication Strategies

Tuesday, November 24, 2009, 5:00pm
Ungar Room 402

Abstract: Patients at risk for complications of influenza are commonly treated with antiviral medications, which however also could be used to control outbreaks. The adamantanes and neuraminidase inhibitors are active against influenza A, but avian influenza (H5N1) is resistant to oseltamivir and swine influenza (H1N1) to the adamantanes (but see postscript). To explore influenza medication strategies (pre-exposure or prophylaxis, post-exposure/pre-symptom onset, and treatment at successive clinical stages) that may affect evolution of resistance (select for resistant strains within or facilitate their spread between hosts), we elaborated a published transmission model and chose parameters from the literature. Then we derived the reproduction numbers of sensitive and resistant strains, peak and final sizes, and time to peak. Finally, we made these results accessible via user-friendly Mathematica notebooks. (Joint work with Rongsong Liu, Dashun Xu, Yiding Yang, and John Glasser)

Professor Sergiy Koshkin
University of Houston-Downtown

Gauge Theory of Faddeev-Skyrme Functionals

Friday, November 20, 2009, 3:30pm
Ungar Room 402

Abstract: We study geometric variational problems for a class of nonlinear sigma-models in quantum field theory. Mathematically, one needs to minimize an energy functional on homotopy classes of maps from closed 3-manifolds into compact homogeneous spaces G/H, similar to the case of harmonic maps. The minimizers are known as Hopfions and exhibit localized knot-like structure. Our main results include proving existence of Hopfions as finite energy Sobolev maps in each (generalized) homotopy class when the target space is a symmetric space. For more general spaces we obtain a weaker result on existence of minimizers in each 2-homotopy class.

Our approach is based on representing maps into G/H by equivalence classes of flat connections. The equivalence is given by gauge symmetry on pullbacks of G-->G/H bundles. We work out a gauge calculus for connections under this symmetry, and use it to eliminate non-compactness from the minimization problem by fixing the gauge.

Dr. Alexander Engström
Miller Research Fellow
University of California, Berkeley

Graph Theoretic Methods in Algebraic Statistics

Thursday, November 12, 2009, 5:00pm
Ungar Room 402

Abstract: First I will review how methods from commutative algebra, for example Gröbner bases and toric ideals, can be used in statistics. Then I will describe two applications of graph theoretic methods in this context: My proof of Sturmfels and Sullivant's conjecture on cut ideals; and the ideals of graph homomorphisms introduced together with Patrik Noren.

Dr. Daniel Ruberman
Brandeis University

Slice Knots and the Alexander Polynomial

Thursday, November 5, 2009, 5:00pm
Ungar Room 402

Abstract: A knot in the 3-sphere is slice if it bounds an embedded disk in the 4-ball. The disk may be topologically embedded, or we may require the stronger condition that it be smoothly embedded; the knot is said to be (respectively) topologically or smoothly slice. It has been known since the early 1980's that there are knots that are topologically slice, but not smoothly slice. These result from Freedman's proof that knots with trivial Alexander polynomial are topologically slice, combined with gauge-theory techniques originating with Donaldson. In joint work with C. Livingston and M. Hedden, we answer the natural question of whether Freedman's result is responsible for all topologically slice knots. We show that the group of topologically slice knots, modulo those with trivial Alexander polynomial, is infinitely generated. The proof uses Heegaard-Floer theory.