DYNAMICAL SYSTEM SEMINAR


Place: Room 303, BS 12 Building, Florida Atlantic University. Time: 2.50 pm



Thursday, March 18th 2010: 2.50 pm, Room: 303 BS-12 (COMPLEX SYSTEMS & BRAIN SCIENCES SEMINAR ROOM)

Larry LIEBOVITCH (Associate Dean, CCS, FAU)

Title: The Mathematics of Psychotherapy

Abstract: We show how three very different problems: two people in a conflict, two genes each inhibiting each other, and the interaction between a therapist and client, can all be represented by similar (but still different) first order coupled nonlinear differential equations. In terms of the therapist-client interaction we ask: What constitutes a good therapeutic relationship? We use the equations to model the dynamics of the emotional interaction between the therapist and client. The solution of these equations shows how the therapeutic endpoint and the dynamics of getting there depends on how the therapist and client influence each other.



Thursday, March 4th 2010th: 2.50 pm, Room: 303 BS-12 (COMPLEX SYSTEMS & BRAIN SCIENCES SEMINAR ROOM)

Iuliana STANCULESCU (NOVA South. East. University)

Title: Approximate Deconvolution Models of Turbulence

Abstract: If the NSE are averaged with a local, spacial, convolution type filter the resulting system is not closed due to the term g * (uu) A deconvolution operator D is one which satisfies:
D(g*u) = approximation of u.

This yields the closure method
g*(uu) ~g*(D(g*u)D(g*u)).

We will review several solutions to the ill-possed deconvolution problem, present the Tikhonov-Lavrentiev deconvolution procedure, and present numerical analysis and numerical experiments with it.



Thursday, February 11th 2010th: 2.50 pm, Room: SE 271

Juan GUTIERREZ (University of Miami)

Title: Genetic Control of Invasive Species: Population Dynamics of the Predator Within

Abstract: This talk is about the dynamics of sexual predators of sorts. As we will see, all present strategies of genetic control of invasive species fall within a small set of categories that take advantage of sexual behavior. We will see that these systems have some predator-prey-like characteristics. strategies. I will present the properties of this family of finite and infinite dimensional ll review several solutions to the ill-possed deconvolution problem, present the Tikhonov- Lavrentiev deconvolution procedure, and present numerical analysis and numerical experiments with it. dynamical systems, and will offer answers to some important biological questions with math.



Thursday, January 28st 2010th: 2.50 pm, Room: SE 271

Hongwei LONG (FAU, Math Dept)

Title: The Connections between Stochastic Differential Equations and Partial Differential Equations.

Abstract: In this talk we shall first briefly discuss some basic concepts such as Brownian motions, stochastic integrals and Ito formula. Then we will mainly focus on establishing the connections between stochastic differential equations and partial differential equations. Finally some examples in financial economics will be provided.


OLDER SEMINAR: Fall 2009:


Thursday, September 10th 2009th: 3.00 pm

Rob VANDERVORST (FAU-Math dept/Vrije Universiteit Amsterdam)


Title: Braids & Dynamics

Abstract: We study the non-autonomous Hamilton equations in the plane. The extended phase space of such a system is 3-dimensional and the orbits may display all kinds of knotting and braiding. We use the topology of braids to construct solutions of the Hamilton equations. This way forcing results (does a given solution force additional solutions?) are obtained. We mention also applications to area-preserving maps.



Thursday, September 24th 2009th: 2.40 pm

Armin FUCHS (Center for Complex Systems and Brain Sciences)


Title: Structure, Dynamics and Function of the Human Brain: Noninvasive Recording Techniques and Realistic Models.

Abstract: Starting from the question: what are the basic requirements to describe and understand a complex system, we give an overview on the modern noninvasive imaging techniques that provide insight into the human brain.s structure and function. We discuss the effects that homogeneous (short-range) and heterogeneous (long-range) connections within the system have on its dynamical behavior based on simple models that include transmission delays. Finally, we show how the folded cortical surface can be incorporated in realistic models of structure, dynamics and function of the human brain.



Thursday, October 8th 2009th: 3.00 pm

Emmanuelle TOGNOLI (Center for Complex Systems and Brain Sciences)


Title: Reading the mysterious language of the brain: a dynamical challenge for the next decade

Abstract: Understanding nature's law, whether in simple manifestations or in complexity, requires a human mind to grasp and interpret patterns and regularities. These patterns can be mapped onto different systems of the human mind: numbers, geometries, equations, words, images, symbols, etc... On all counts, the human brain is a complex system, and it poses so many challenges that some thinkers questioned whether it could ever understand itself. Rather than avoiding this inherent paradox, we embraced it and enquired about the brain that understands prior to asking about the brain to be understood. Brain function can be framed as a spatio-temporal problem. Its neurons are fixed and functionally arranged as a result of philogeny and ontogeny. Its activity is ever-changing in time, and I will show how distinctly ephemeral are brain functional patterns as compared to those from systems that loose or lack complexity. For the typical human observer, space and time are perceived as different currencies of nature. Our recent efforts went into exposing brain activity in time and cortical surface simultaneously; a 4-dimensional space at the very least. Natural human perception could not be relied upon; it is bounded to a maximum of 3-dimensions. To circumvent this ceiling, the trick was to understand how order emerges from the human mind and to reroute spatial and temporal information to perceptual channels that could run in concert. It gave rise to a 4-dimensional colorimetric visualization of spatio-temporal data. The technique opened up the possibility to read functional states of the brain from continuous records of its activity; and also to come to grip with testing theories of brain self-organization. The upcoming challenge is to decipher the mysterious language of the brain and maybe someday to establish fundamental laws of the human minds.



Thursday, October 22th 2009th: 3.00 pm

SPECIAL SEMINAR: Ram MOHAPATRA (Central Florida University)


Title: On Epidemiological models with mutating pathogens

Abstract: In this talk we shall discuss some epidemiological models for the transmission of a pathogen that can mutate in a host to create a second infectious mutant strain. Explicit formulas for the reproductive number on an epidemic based on the local stability of the infection-free equilibrium will be mentioned. We shall also analyze the existence and stability of the boundary equilibrium and the endemic equilibrium. We shall talk about the global stability of the boundary equilibrium. Finally we shall show that under certain circumstances there is Hopf bifurcation where the endemic equilibrium loses its stability, and periodic solutions appear. Some numerical simulations to illustrate Hopf bifurcation will be considered. Finally recent results on application of Homotopy Analysis Method to epidemiological models will be mentioned.



Thursday, October 29th 2009th: 2.50 pm

Robert ROUSSARIE (Institut Mathematiques de Bourgogne)

Title: Bifurcation Theory For Planar Vector Fields

Abstract: In contrast with the situation in higher dimensions, a bifurcation theory for multi-parameter families of vector fields was developed in dimension 2 until some point, especially on the plane. First, I want to recall some basic notions as for instance what is a versal unfolding and make precise the ultimate goal for the theory: that is to obtain all possible versal unfoldings and to find a good description of generic k-parameter families in terms of these versal unfoldings. Next I shall review some important results obtained up to now and I shall also mention some open questions. Finally I shall give an idea about the methods we use: normal forms, rescaling and desingularization, asymptotics.



Thursday, November 19th 2009th: 3 pm

Silke DODEL (Center for Complex Systems and Brain Sciences)

Title: Objective dynamical measures of team coordination and performance

Abstract: team is more than just a group of people. But how can team coordination and performance be measured? A team consisting of one excellent and several poor members may score reasonably well by conventional standards, but may not in terms of team coordination. We developed novel measures of team performance and team coordination that overcome limitations of current team performance measures, such as being subjective and ignoring dynamic team processes. By using concepts from theoretical physics and dynamical systems theory we represent team dynamics geometrically as a manifold. The deviation of the actual team trajectory from the optimal manifold provides an objective measure of team performance, while the direction of the deviation informs about the nature of the performance deficiency. By expanding this approach we in addition assess the evolution of team coordination over time, both for the team as a whole and for all pairs of individual team members to reveal dynamic coordination patterns in teams.



Thursday, December 3rd 2009th: 3 pm

Marcus FONTAINE (FAU, Math Dept)

Title: Period Doubling Bifurcations & Reduction Techniques

Abstract: Period doubling bifurcations occur in many models, including overcompensation in Population Dynamics, for example. We analyze period doubling bifurcations for 1-dimensional maps. We then illustrate how dynamics in Multi-dimensional systems can be reduced to 1-dimension, with techniques involving center manifolds or Lyapunov-Schmidt reduction, where the 1-dimensional analysis can now be applied.