**Date:** Mar 13, 2020

**Speaker:**
Shravan Veerapaneni, University of Michigan, Ann Arbor

**Title:** Fast solvers for simulating particulate flows in complex geometries

**Abstract:**
From blood flow to subsurface flows, particulate flows are ubiquitous. Direct numerical simulations of dense, rigid or deformable particle suspensions in viscous fluids are extremely challenging yet critically important to bring insights into their macro-scale flow behavior. In this talk, I will present recent advances made by our group in overcoming several computational bottlenecks such as accurate evaluation of nearly-singular integrals, periodization schemes for complex geometries and robust collision resolution algorithms. Applications in the design of microfluidic chips, shape optimization and electrohydrodynamics of vesicles will be discussed.

**Date:** Feb 28, 2020

**Speaker:** Samuli Siltanen, University of Helsinki, Finland

**Title:**
Classifying stroke from electric boundary data by nonlinear Fourier analysis

**Abstract:**
Abstract: Stroke is a leading cause of death all around the world. There are two main types of stroke: ischemic (blood clot preventing blood flow to a part of the brain) and hemorrhagic (bleeding in the brain). The symptoms are the same, but treatments very different. A portable “stroke classifier" would be a life-saving equipment to have in ambulances, but so far it does not exist. Electrical Impedance Tomography (EIT) is a promising and harmless imaging method for stroke classification. In EIT one attempts to recover the electric conductivity inside a domain from electric boundary measurements. This is a nonlinear and ill-posed inverse problem. The so-called Complex Geometric Optics (CGO) solutions have proven to be a useful computational tool for reconstruction tasks in EIT. A new property of CGO solutions is presented, showing that a one-dimensional Fourier transform in the spectral variable provides a connection to parallel-beam Xray tomography of the conductivity. One of the consequences of this “nonlinear Fourier slice
theorem” is a novel capability to recover inclusions within inclusions in EIT. In practical imaging, measurement noise causes strong blurring in the recovered profile functions. However, machine learning algorithms can be combined with the nonlinear PDE techniques in a fruitful way. As an example, simulated strokes are classified into hemorrhagic and ischemic using EIT measurements.

**Date:** Feb 21, 2020

**Speaker:** Eric Bonnetier, Institut Fourier Grenoble,

**Title:** Homogenization of the Poincare Neumann operator

**Abstract:**
The Neumann Poincar\'e operator is an integral operator that allows the representation of the solutions to elliptic
PDE's with piecewise constant coefficients using layer potentials. Its spectral properties are of interest in the study
of plasmonic resonances of metallic particles.
We discuss the spectrum of that integral operator, when one considers a periodic distribution of inclusions made of
metamaterials in a dielectric background medium. We show that under the assumption that the inclusions are fully
embedded in the periodicity cells, the limiting spectra of periodic NP operators is composed of a Bloch spectrum,
and of a boundary spectrum associated with eigenfunctions which concentrate a part of their energy near the boundary.
This is joint work with Charles Dapogny and Faouzi Triki.

**Date:** Feb 14, 2020

**Speaker:** Hiroshi Takeuchi, Chubu University, Japan

**Title:** Application of persistent homology to granular materials and sampled dynamical systems

**Abstract:**
Persistent homology is a tool describing the shape of data. In this talk, we apply persistent homology to two topics; granular materials and sampled dynamical systems. A granular material is a set of macroscopic particles, such as sand, nuts, or coffee beans. The simplest model of granular materials is monodisperse sphere packings. We utilize persistent homology to capture the change of topological configurations in the crystallization of sphere packings.
In the second application, we focus on a sampled map, which is a finite subset of a map. A notable example is that the map is a discrete dynamical system, then a sampled map is a sampling from the dynamical system. Our motivation is to retrieve topological information of the dynamical system only from the finite sampling data. Persistent homology can extract the robust topological changes in the underlying map, and the cycles of the homology generators provide the model of the underlying map.

**Date:** Feb 7, 2020

**Speaker:** Zin Arai, Chubu University

**Title:** Period doubling bifurcations from complex and algebraic point of view

**Abstract:**
It is well known in one-dimensional dynamical systems that the period doubling
bifurcations plays an essential role in the creation of chaos. In this
talk, we will
see that this is also the case for higher dimensional systems by studying the
monodromy representation of the system and the dynamical zeta function.
This is a first step to understand the mysterious topology of the higher
dimensional analog of the Mandelbrot set.

**Date:** Dec 6, 2019

**Speaker:** Catalin Turc, NJIT

**Title:** Optimized Schwarz Methods for the iterative solution of quasiperiodic Helmholtz transmission problems in layered media

**Abstract:**
We present an Optimized Schwarz Domain Decomposition Methods applied to Helmholtz transmission problems in periodic layered media. Unlike the classical domain decomposition approach that relies on exchange of Robin data on the subdomain boundaries, we incorporate instead transmission operators that are approximations of Dirichlet-to-Neumann (DtN) operators. The latter approximations, in turn, is obtained via shape perturbation series. The Robin-to-Robin (RtR) operators that are the building blocks of Domain Decomposition Methods are expressed via boundary integral equation formulations that are shown to be robust for all frequencies, including the challenging Wood frequencies.
We use Nyström discretizations of quasi-periodic boundary integral operators to construct high-order approximations of RtR. Based on the premise that the quasi-optimal transmission operators should act like perfect transparent boundary conditions, we construct an approximate LU factorization of the tridiagonal QO DD matrix associated with periodic layered media, which is then used as a double sweep preconditioner. We present a variety of numerical results that showcase the effectiveness of the sweeping preconditioners for the iterative solution of Helmholtz transmission problems in periodic layered media.
Joint work with David Nicholls (UIC) and Carlos Perez Arancibia (PUC Chile)

**Date:** Nov 22, 2019

**Speaker:** Isaac Harris, Purdue University

**Title:** Direct Sampling Algoritmis in Inverse Scattering

**Abstract:**
In this talk, we will discuss a recent qualitative imaging method referred to as the Direct Sampling Method for inverse scattering. This method allows one to recover a scattering object by evaluating an imaging functional that is the inner-product of the far-field data and a known function. It can be shown that the imaging functional is strictly positive in the scatterer and decays as the sampling point moves away from the scatterer. The analysis uses the factorization of the far-field operator and the Funke-Hecke formula. This method can also be shown to be stable with respect to perturbations in the scattering data. We will discuss the inverse scattering problem for both acoustic and electromagnetic waves.

**Date:** Nov 15, 2019

**Speaker:** Shawn Walker, Louisiana State University

**Title:** The Uniaxially Constrained Q-tensor Model for Nematic Liquid Crystals

**Abstract:**
We consider the one-constant Landau-de Gennes (LdG) model for nematic liquid crystals with traceless tensor field Q as the order parameter that seeks to minimize a Dirichlet energy plus a double well potential that confines the eigenvalues of Q (examples/applications will be described). Moreover, we constrain Q to be uniaxial, which involves a rank-1 constraint. Building on similarities with the one-constant Ericksen energy, we propose a structure-preserving finite element method for the computation of equilibrium configurations. We prove stability and consistency of the method without regularization, and $\Gamma$-convergence of the discrete energies towards the continuous one as the mesh size goes to zero. We also give a monotone gradient flow scheme to find minimizers. We illustrate the method's capabilities with several numerical simulations in two and three dimensions including non-orientable line fields. In addition, we do a direct comparison between the standard LdG model, and the uniaxially constrained model.

**Date:** Nov 8, 2019

**Speaker:** Jingni Xiao, Rutgers University

**Title:** Corner scattering and some applications

**Abstract:**
We consider time-harmonic medium or source scattering. We examine the effect of never-trivial-scattering due to the appearance of corners at the support of the medium inhomogeneity or the source. In particular, interior transmission eigenfunctions in a cornered domain can not be extended into the neighborhood as an incident wave field. Some applications like inverse scattering on shape determination will also be discussed.

**Date:** Oct 25, 2019

**Speaker:** Alex Blumenthal, University of Maryland

**Title:** Lyapunov exponents for small random perturbations of predominantly hyperbolic volume-preserving diffeomorphisms, including the Standard Map

**Abstract:**
An outstanding problem in smooth ergodic theory is the estimation from below of Lyapunov exponents for maps which exhibit hyperbolicity on a large but non-invariant subset of phase space, e.g. the Chirikov standard map or Henon map families. It is notoriously difficult to show that Lyapunov exponents actually reflect the predominant hyperbolicity in the system, due to cancellations caused by the switching of stable and unstable directions in those parts of phase space where hyperbolicity is violated. In this talk I will discuss the inherent difficulties of the above problem, and will discuss recent results when small random perturbations are introduced at every time-step. In this case, we show that for a large class of predominantly hyperbolic systems in two dimensions, the top Lypaunov exponent is large in proportion to the strength of the predominant hyperbolicity in the system. Our results apply to the standard map with large coefficient. This work is joint with Lai-Sang Young and Jinxin Xue.

**Date:** Oct 18, 2019

**Speaker:** Tianhao Zhang, Rutgers University

**Title:** A constructive proof of the Cauchy-Kovalevskaya theorem with applications to validated numerics.

**Abstract:**
In this talk, I will present a constructive proof of the Cauchy-Kovalevskaya theorem for ODEs. The proof is motivated by a validated numerics technique called the radii polynomial approach commonly used for polynomial ODEs. I will introduce the basic aspects of this approach and then show how we extend it to the analytic case and apply it to prove the classical Cauchy-Kovalevskaya theorem. This is joint work with Shane Kepley.

**Date:** Oct 11, 2019

**Speaker:** William Cuello, Rutgers University

**Title:** Single and Multispecies Persistence in the Face of Environmental Uncertainty

**Abstract:**
I will be presenting my work on the analyses of single and multispecies systems in fluctuating environments. For the first half of the presentation, I will present my work on predicting average, germination rates of 10 bet-hedging Sonoran Desert annuals; here, annuals hedge their bets by keeping a fraction of their seeds dormant to buffer against uncertain amounts of annual rainfall. For the second half, I will present the mathematical framework I have developed for analyzing classes of multispecies, stochastic models. This framework provides a step-by-step process in which one can determine whether a system of species will stochastically persist.

**Date:** Oct 4, 2019

**Speaker:** Elena Queirolo, Vrije University

**Title:** Hopf bifurcation in PDE

**Abstract:**
In this talk we will use validated numerics to prove the existence of a Hopf bifurcation in the Kuramoto-Sivashinky PDE.

Validated numerics, in particular the radii polynomial approach, allows to prove the existence of a solution to a given problem in the neighborhood of a numerical approximation. A known use of this technique is for branch following in parameter dependent ODEs. In this talk, we will consider periodic solutions of polynomial ODEs.

With a blow up approach, we can rewrite the original ODE into a new system that undergoes a saddle node bifurcation instead of a Hopf bifurcation, thus avoiding the singularity of the periodic solution. We can then prove the existence of a Hopf bifurcation in the original system with a combination of analytical and validated numeric results.

To conclude, we will apply the same techniques in the Kuramoto-Sivashinky case, thus demostrating the flexibility of this approach.

**Date:** Sep 27, 2019

**Speaker:** Andreas Kirsch, KIT

**Title:** A Radiation Condition for the Scattering by Locally Perturbed Periodic Layers

**Abstract:**
Scattering of time-harmonic waves from periodic structures at some fixed real-valued wave number becomes analytically difficult whenever there arise surface waves: These non-zero solutions to the homogeneous scattering problem physically correspond to modes propagating along the periodic structure and clearly imply non-uniqueness of any solution to the scattering problem. In this talk, I consider a medium that is defined in the upper two-dimensional half-space by a penetrable and periodic contrast. We formulate a proper radiation condition which is motivated by the limiting absorption principle; that is, the solution is the limit of a sequence of unique solutions for artificial complex-valued wave numbers tending to the above-mentioned real-valued wave number. By the Floquet-Bloch transform we first reduce the scattering problem to a finite-dimensional one that is set in the linear space spanned by all surface waves. In this space, we then compute explicitly which modes propagate along the periodic structure to the left or to the right. This finally yields a representation for our limiting absorption solution which leads to a proper extension of the well known upward propagating radiation condition. Finally, we briefly consider the case when the periodic refractive index is perturbed locally.

**Date:** Sep 20, 2019

**Speaker:** Brittany Hamfeldt, NJIT

**Title:** Generalised finite difference methods for fully nonlinear elliptic equations

**Abstract:**
The introduction of viscosity solutions and the Barles-Souganidis convergence framework have allowed for considerable progress in the numerical solution of fully nonlinear elliptic equations. We describe a framework for constructing convergent generalised finite difference approximations for a large class of nonlinear elliptic operators. These approximations are defined on unstructured point clouds, which allows for computation on non-uniform meshes and complicated geometries. Because the schemes are monotone, they fit within the Barles-Souganidis convergence framework and can serve as a foundation for higher-order filtered methods. We present computational results for several examples including problems posed on random point clouds, examples incorporating automatic mesh adaptation, non-continuous surfaces of prescribed Gaussian curvature, Monge-Ampere equations arising in optimal transportation, and Monge-Ampere type equations on the sphere.

**Date:** Sep 13, 2019

**Speaker:** Shane Kepley, Rutgers University

**Title:** Computing linear extensions of partial orders subject to algebraic constraints

**Abstract:**
Switching systems have been extensively used for modeling the dynamics of gene regulation. This is partially due to the natural decomposition of phase space into rectangles on which the dynamics can be completely understood. Recent work has shown that the parameter space can also be decomposed into ``nice'' subsets. However, computing these subsets, or even determining if they are empty turns out to be a difficult problem.

We will show that computing a parameter space decomposition is equivalent to computing all linear extensions of a certain poset. The elements of this poset are polynomials and this structure induces additional algebraic constraints on the allowable linear extensions. We will describe an algorithm for efficiently solving this problem when the polynomials are linear. A more general class of polynomials can also be handled efficiently through a transformation which reduces to the linear case. Finally, we present several open problems and conjectures which arise when one generalizes this problem to subsets of arbitrary polynomial rings.

**Date:** April 26, 2019

**Speaker:** Harbir Antil, George Mason University

**Title:** Fractional PDEs: Optimal Control and Applications

**Abstract:**
Fractional calculus and its application to anomalous transport has recently received a tremendous amount of attention. In these studies, the anomalous transport (of charge, tracers, fluid, etc.) is presumed attributable to long-range correlations of material properties within an inherently complex, and in some cases self-similar, conducting medium. Rather than considering an exquisitely discretized (and computationally explosive) representation of the medium, the complex and spatially correlated heterogeneity is represented through reformulation of the PDE governing the relevant transport physics such that its coefficients are, instead, smooth but paired with fractional-order space derivatives.
This talk will give an introduction to fractional diffusion. We will describe how to incorporate nonhomogeneous boundary conditions in fractional PDEs. We will cover from linear to quasilinear fractional PDEs. New notions of exterior optimal control and optimization under uncertainty will be presented. We will conclude the talk with an approach that allows the fractional exponent to be spatially dependent. This has enabled us to define novel Sobolev spaces and their trace spaces. Several applications in: imaging science, quantum random walks, geophysics, and manifold learning (data analysis) will be discussed.

**Date:** April 10, 2019

**Speaker:** Michael Levitin, University of Reading, UK

**Title:** Sharp eigenvalue asymptotics for Steklov problem on curvilinear polygons

**Abstract:**
I will discuss a work in progress (joint with Leonid Parnovski, Iosif Polterovich and David Sher) on Steklov (or Dirichlet-to-Neumann map) eigenvalue asymptotics for curvilinear polygonal domains in R^2. The results are quite unexpected, and the asymptotics depends non-trivially on the arithmetic properties of the angles of the polygon. There are also connections to classical problems of hydrodynamics (the sloping beach problem and the sloshing problem) and to the Laplacian on quantum graphs.

**Date:** February 15, 2019

**Speaker:** Ricardo Nochetto, University of Maryland, College Park

**Title:** Thermally Actuated Bilayer Plates

**Abstract:**
We present a simple mathematical model of polymer bilayers that
undergo large bending deformations when actuated by non-mechanical
stimuli such as thermal effects. The model consists of a
nonlinear fourth order problem with a pointwise isometry
constraint, which we discretize with either Kirchhoff quadrilaterals
or discontinuous Galerkin methods.
We prove $\Gamma$-convergence of the discrete model and propose an
iterative method that decreases its energy and leads to stationary
configurations. We investigate performance, as well as reduced
model capabilities, via several insightful numerical experiments
involving large (geometrically nonlinear) deformations. They include
the folding of several practically useful compliant structures
comprising of thin elastic layers. This work is joint with S. Bartels,
A. Bonito, and D. Ntogkas.

**Date:** February 1, 2019

**Speaker:** Heather Harrington, Oxford University

**Title:** Comparing models and data using computational algebraic geometry and topology.

**Abstract:**
I will overview my research for a very general math audience. I will start with motivation of the
biological problems we have explored, such as tumor-induced angiogenesis (the growth of blood vessels to
nourish a tumor), as well as signaling pathways involved in the dysfunction of cancer (sets of molecules that
interact that turn genes on/off and ultimately determine whether a cell lives or dies).
Both of these biological problems can be modeled using differential equations. The challenge with analyzing
these types of mathematical models is that the rate constants, often referred to as parameter
values, are difficult to measure or estimate from available data.

I will present mathematical methods we have developed to enable us to compare mathematical models with experimental data. Depending on the type of data available, and the type of model constructed, we have combined techniques from computational algebraic geometry and topology, with statistics, networks and optimization to compare and classify models without necessarily estimating parameters. Specifically, I will introduce our methods that use computational algebraic geometry (e.g., Grobner bases) and computational algebraic topology (e.g., persistent homology). I will present applications of our methodology on datasets involving cancer. Time permitting, I will conclude with our current work for analyzing spatio-temporal datasets with multiple parameters using computational algebraic topology. Mathematically, this is studying a module over a multivariate polynomial ring, and finding discriminating and computable invariants.

**Date:** December 11, 2018

**Speaker:** Yasumasa Nishiura, Advanced Institute for Materials Research, Tohoku University

**Title:** What is a good mathematical descriptor for the toughness of
heterogeneous materials

**Abstract:**
One of the dreams of materials scientists is to make a novel materials
through the design of atomistic scale, however most of the modern
composite materials is very heterogeneous in order to make it strong
via network structure like epoxy resin matrix with carbon fibers used in
the aircraft. Those are far from crystal structure nor completely random
so that it is not apriori clear what type of mathematical concepts is
appropriate to describe it, especially medium-range structure. As for
the static profile, recent statistical methods as well as topological
approach TDA clarify some aspects of it.
On the other hand, the performance of the materials, especially its
dynamic robusness against mechanical stress remains open and still
heavily depends on trials and errors in the laboratories. The difficulty
lies in that, firstly the lack of good macroscopic mathematical model to
describe dynamical processes, secondly it is not clear how to implement
the microscopic heterogeneity into the macroscopic model, thirdly how to
extract an appropriate mathematical descriptor that can measure and
predict the strength of it and even allows us to design novel materials.
I would like to present a case study in this direction in the context of
cracking phenomena for brittle materials. The stage is still early phase, however
it suggests many interesting questions and challenge not only for for
materials scientists but for mathematicians.

**Date:** November 16, 2018

**Speaker:** Alexander Vladimirsky, Cornell University

**Title:** Surveillance-Evasion games under uncertainty

**Abstract:**
Adversarial path planning problems are important in robotics applications and in modeling the behavior of humans in dangerous environments. Surveillance-Evasion (SE) games form an important subset of such problems and require a blend of numerical techniques from multiobjective dynamic programming, game theory, numerics for Hamilton-Jacobi PDEs, and convex optimization. We model the basic SE problem as a semi-infinite zero-sum game between two players: an Observer (O) and an Evader (E) traveling through a domain with occluding obstacles.
O chooses a pdf over a finite set of predefined surveillance plans, while E chooses a pdf over an infinite set of trajectories that bring it to a target location. The focus of this game is on "E's expected cumulative exposure to O", and we have recently developed an algorithm for finding the Nash Equilibrium open-loop policies for both players. I will use numerical experiments to illustrate algorithmic extensions to handle multiple Evaders, moving Observes, and anisotropic observation sensors. Time permitting, I will also show preliminary results for a very large number of selfish/independent Evaders modeled via Mean Field Games.
Joint work with M.Gilles, E.Cartee, and REU-2018 participants.

**Date:** November 2, 2018

**Speaker:** Abner Salgado, University of Tennessee, Knoxville

**Title:** Regularity and rate of approximation for obstacle problems for a class of integro-differential operators

**Abstract:**
We consider obstacle problems for three nonlocal operators:

A) The integral fractional Laplacian

B) The integral fractional Laplacian with drift

C) A second order elliptic operator plus the integral fractional
Laplacian

For the solution of the problem in Case A, we derive regularity results
in weighted Sobolev spaces, where the weight is a power of the distance
to the boundary. For cases B and C we derive, via a Lewy-Stampacchia
type argument, regularity results in standard Sobolev spaces. We use
these regularity results to derive error estimates for finite element
schemes. The error estimates turn out to be optimal in Case A, whereas
there is a loss of optimality in cases B and C, depending on the order
of the integral operator.

**Date:** October 26, 2018

**Speaker:** Bill Kalies, Florida Atlantic University

**Title:** Order Theory in Dynamics

**Abstract:**
Recurrent versus gradient-like behavior in global dynamics
can be characterized via a surjective lattice homomorphism between
certain bounded, distributive lattices, that is, between attracting blocks
(or neighborhoods) and attractors. In this lecture we explain the basic
order and lattice theory for dynamical systems which lays a foundation
for a computational theory for dynamical systems that focuses on Morse
decompositions and index lattices. We build combinatorial order-theoretic
models for global dynamics. We give computational examples that
illustrate the theory for both maps and flows.

**Date:** October 24, 2018, 5:00-6:00pm

**Room:** Hill 005

**Speaker:** Prof. Wojciech Chacholski, Department of Mathematics, KTH

**Title:** What is persistence

**Abstract:**
It is not surprising that different units and scales are used to measure different phenomena. So why the Gromov-Hausdorff and bottleneck distances are the only one used to measure inputs and outcomes of topological data analysis applied to a variety of different data sets? My aim is to explain and illustrate a new approach to persistence. I will present both mathematical and real life data examples illustrating effectiveness of our approach to improve various classification tasks.

**Date:** October 19, 2018

**Speaker:** Francisco Sayas, University of Delaware

**Title:** Waves in viscoelastic solids

**Abstract:**
I will first explain a transfer function based framework collecting well-known models of wave propagation in viscoelastic solids, fractional derivative extensions, and their couplings.
I will briefly explain the associated Laplace domain stability results and their time domain counterparts, as well as how they are affected by finite element discretization in space.
Finally, I will discuss a semigroup approach to a non-strictly diffusive Zener model. This is joint work with Tom Brown, Shukai Du, and Hasan Eruslu.

**Date:** October 12, 2018

**Speaker:** Vladimir Itskov, The Pennsylvania State University

**Title:** Directed complexes, sequence dimension and inverting a neural network.

**Abstract:**
What is the embedding dimension, and more generally, the geometry of a set of sequences? This problem arises in the context of neural coding and neural networks. Here one would like to infer the geometry of a space that is measured by unknown quasiconvex functions. A natural object that captures all the inferable geometric information is the directed complexes (a.k.a. semi-simplicial sets). It turns out that the embedding dimension as well as some other geometric properties of data can be estimated from the homology of an associated directed complex. Moreover each such directed complex gives rise to a multi-parameter filtration that provides a dual topological description of the underlying space. I will also illustrate these methods in the neuroscience context of understanding the "olfactory space".

**Date:** September 28, 2018

**Speaker:** Carina Curto, The Pennsylvania State University

**Title:** Graph rules for inhibitory network dynamics

**Abstract:**
Many networks in the nervous system possess an abundance of inhibition, which serves to shape and stabilize neural dynamics. The neurons in such networks exhibit intricate patterns of connectivity, whose structure controls the allowed patterns of neural activity. In this work, we examine inhibitory threshold-linear networks whose dynamics are dictated by an underlying directed graph. We develop a set of parameter-independent graph rules that enable us to predict features of the dynamics from properties of the graph. These rules provide a direct link between the structure and function of these networks, and provides new insights into how connectivity may shape dynamics in real neural circuits.

**Date:** December 1, 2017

**Speaker:** Harbir Antil, George Mason University

**Title:** Fractional Operators with Inhomogeneous Boundary Conditions: Analysis, Control, and Discretization

**Abstract:**
In this talk we introduce new characterizations of spectral fractional Laplacian to incorporate non homogeneous Dirichlet and Neumann boundary conditions. The classical cases with homogeneous boundary conditions arise as a special case. We apply our definition to fractional elliptic equations of order $s \in (0,1)$ with nonzero Dirichlet and Neumann boundary conditions. Here the domain $\Omega$ is assumed to be a bounded, quasi-convex.

**Date:** November 3, 2017

**Speaker:** Marcio Gameiro, University of Sao Paulo at Sao Carlos, Brazil

**Title:** Rigorous Multi-parameter Continuation of Solutions of Differential Equations

**Abstract:**
We present a rigorous multi-parameter continuation method to compute solutions
of differential equations depending on parameters. The method combines classical
numerical methods, analytic estimates and the uniform contraction principle to prove
the existence of solutions of nonlinear differential equations. The method is applied to
the computation of equilibria for the Cahn-Hilliard equation and periodic solutions
of the Kuramoto-Sivashinsky equation.

**Date:** September 22, 2017

**Speaker:** Qi Wang, University of South Carolina

**Title:** Energy quadratization strategy for numerical approximations of nonequilibrium models

**Abstract:**
There are three fundamental laws in equilibrium thermodynamics. But, what are the laws in nonequilibrium thermodynamics that guides the development of theories/models to describe nonequilibrium phenomena? Continued efforts have been invested in the past on developing a general framework for nonequilibrium thermodynamic models, which include Onsager's maximum entropy theory, Prigogine's minimum entropy production rate theory, Poisson bracket formulation of Beris and Edwards, as well as the GENERIC formalism promoted by Ottinger and Grmela. To some extent, they are equivalent and all give practical means to develop nonequilibrium dynamic models. In this talk, I will focus on the Onsager approach, termed the Generalized Onsager Principle (GOP). I will review how one can derive thermodynamic and generalized hydrodynamic models using the generalized Onsager principle coupled with the variational principle. Then, I will discuss how we can exploit the mathematical structure of the models derived using GOP to design structure and property preserving numerical approximations to the governing system of partial differential equations. Since the approach is valid near equilibrium as pointed it out by Onsager, an energy quadratization strategy is proposed to arrive linear numerical schemes. This approach is so general that in principle we can use it to any nonequilibrium model so long as it has the desired variational and dissipative structure. Some numerical examples will be given to illustrate the usefulness of this approach.

**Date:** April 20, 2017

**Speaker:** Michael Neilan, University of Pittsburgh

**Title:** Discrete theories for elliptic problems in non--divergence
form

**Abstract:** In this talk, two discrete theories for elliptic problems in
non-divergence form are presented. The first, which is applicable to problems
with continuous coefficients and is motivated by the strong solution concept,
is based on discrete Calderon-Zygmund-type estimates. The second theory relies
on discrete Miranda-Talenti estimates for elliptic problems with discontinuous
coefficients satisfying the Cordes condition. Both theories lead to simple,
efficient, and convergent finite element methods. We provide numerical
experiments which confirm the theoretical results, and we discuss possible
extensions to fully nonlinear second order PDEs.

**Date:** March 3, 2017

**Speaker:** Ridgway Scott, University of Chicago

**Title:** Electron correlation in van der Waals interactions

**Abstract:** We examine a technique of Slater and Kirkwood which provides an
exact resolution of the asymptotic behavior of the van der Waals attraction
between two hydrogens atoms. We modify their technique to make the problem
more tractable analytically and more easily solvable by numerical
methods. Moreover, we prove rigorously that this approach provides an exact
solution for the asymptotic electron correlation. The proof makes use of
recent results that utilize the Feshbach-Schur perturbation technique. We
provide visual representations of the asymptotic electron correlation
(entanglement) based on the use of Laguerre approximations.We also describe an
a computational approach using the Feshbach-Schur perturbation and
tensor-contraction techniques that make a standard finite difference approach
tractable.

**Date:** April 22, 2016

**Speaker:** Guillaume Bal, Columbia University

**Title:** Boundary control in transport and diffusion equations

**Abstract:** Consider a prescribed solution to a diffusion equation in a
small domain embedded in a larger one. Can one (approximately) control such a
solution from the boundary of the larger domain? The answer is positive and
this form of Runge approximation is a corollary of the unique continuation
property (UCP) that holds for such equations. Now consider a (phase space,
kinetic) transport equation, which models a large class of scattering
phenomena, and whose vanishing mean free path limit is the above diffusion
model. This talk will present positive as well as negative results on the
control of transport solutions from the boundary. In particular, we will show
that internal transport solutions can indeed be controlled from the boundary
of a larger domain under sufficient convexity conditions. Such results are not
based on a UCP. In fact, UCP does not hold for any positive mean free path
even though it does apply in the (diffusion) limit of vanishing mean free
path. These controls find applications in inverse problems that model a large
class of coupled-physics medical imaging modalities. The stability of the
reconstructions is enhanced when the answer to the control problem is
positive.

**Date:** April 8, 2016

**Speaker:** John Sylvester, University of Washington

**Title:** Evanescence, Translation, and Uncertainty Principles in the
Inverse Source Problem

**Abstract:** The inverse source problem for the Helmholtz equation (time
harmonic wave equation) seeks to recover information about a radiating source
from remote observations of a monochromatic (single frequency) radiated wave
measured far from the source (the far field). The two properties of far fields
that we use to deduce information about shape and location of sources depend
on the physical phenomenon of evanescence, which limits imaging resolution to
the size of a wavelength, and the formula for calculating how a far field
changes when the source is translated. We will show how adaptations of
"uncertainty principles", as described by Donoho and Stark [1] provide a very
useful and simple tool for this kind of analysis.

**Date:**March 24, 2016

**Speaker:** Qi Wang , Interdisciplinary Mathematics Institute and
NanoCenter at University of South Carolina

**Title:** Onsager principle, generalized hydrodynamic theories and
energy stable numerical schemes

**Abstract:** In this talk, I will discuss the Onsager principle for
nonequilibrium thermodynamics and present the generalized Onsager principle
for deriving generalized hydrodynamic theories for complex fluids and active
matter. For closed matter systems, the generalized Onsager principle combines
variational principle with the dissipative property of the system to give a
hydrodynamic system that dissipates the total energy. I will illustrate the
idea using a few examples in complex fluids. For the hydrodynamic system of
equations derived from the generalized Onsager principle, dissipation property
preserving numerical schemes can be devised , known as energy stable
schemes. These schemes are unconditional stable in time. Several applications
of generalized hydrodynamic theories to active matter systems, like cell
migration on solid substrates and cytokinesis of animal cells will be
presented.

**Date:** February 26, 2016

**Speaker:** Andrea Bonito, Texas A&M University

**Title:** Bilayer Plates: From Model Reduction to Gamma-Convergent
Finite Element Approximation

**Abstract:** The bending of bilayer plates is a mechanism which allows for
large deformations via small externally induced lattice mismatches of the
underlying materials. Its mathematical modeling consists of a geometric
nonlinear fourth order problem with a nonlinear pointwise isometry constraint
and where the lattice mismatches act as a spontaneous curvature. A gradient
flow is proposed to decrease the system energy and is coupled with finite
element approximations of the plate deformations based on Kirchhoff
quadrilaterals. In this talk, we give a general overview on the model
reduction procedure, discuss to the convergence of the iterative algorithm
towards stationary configurations and the Gamma-convergence of their finite
element approximations. We also explore the performances of the numerical
algorithm as well as the reduced model capabilities via several insightful
numerical experiments involving large (geometrically nonlinear)
deformations. Finally, we briefly discuss applications to drug delivery, which
requires replacing the gradient flow relaxation by a physical flow.

**Date:** February 26, 2016

**Speaker:** Lou Kondic, New Jersey Institute of Technology

**Title:** Force networks in particulate-based systems: persistence,
percolation, and universality

**Abstract:** Force networks are mesoscale structures that form
spontaneously as particulate-based systems (such as granulars, emulsions,
colloids, foams) are exposed to shear, compression, or impact. The
presentation will focus on few different but closely related questions
involving properties of these networks:

(i) Are the networks universal, with their properties independent of those of
the underlying particles?

(ii) What are percolation properties of these networks, and can we use the
tools of percolation theory to explain their features?

(iii) How to use topological tools, and in particular persistence approach to
quantify the properties of these networks?

The presentation will focus on the results of molecular dynamics/discrete
element simulations to discuss these questions and (currently known) answers,
but I will also comment and discuss how to relate and apply these results to
physical experiments.