University of Georgia
Department of Mathematics

Seminar Schedule
March 15 - 19, 2004

All Seminars are held in Boyd Graduate Studies Bldg. unless otherwise noted.

MONDAY, March 15, 2004

VIGRE - Algebra
2:30p.m., Room 410
Speaker: Daniel Nakano, University of Georgia
Title of talk: Support varieties for Weyl modules

Topology
2:30p.m., Room 322
Speaker:
Nathan.Habegger, University of Nantes, France
Title of talk: Some Remarks on the Poincare Conjecture
Abstract:
A pseudo-handlebody is a 3-manifold M for which the fundamental group of the boundary of M surjects onto
the fundamental group of M modulo the intersection of the lower central series. The Poincare Conjecture can be reformulated in terms of splittings of S^3 into pseudo-handlebodies. Work of Scharleman-Thompson on Heegard splittings can be used to further simplify the reformulation. One can also introduce the notion of the Milnor invariant (higher linking numbers) and give a reformulation in terms of these.

Faculty and Graduate Social
3:00p.m., Room 409
Coffee, Tea, Cookies

Lie Theory
3:30p.m., Room 303
No Meeting this week

Stochastic Processes
3:30p.m., Room 410
No Meeting this week

TUESDAY, March 16, 2004

VIGRE Graduate Student Seminar
2:00p.m., Room 304
Speaker:
Ed Azoff, University of Georgia
Title of talk:
Dynamics in [0,1)
Abstract: Let E be an equivalence relation on a set X. A _uniformization_ of E is a subset of X which meets each equivalence class determined by E in precisely one point. Finding "nice" uniformizations is a common mathematical goal. For example, Jordan canonical forms solve the uniformization problem for the relation of similarity on complex matrices.
After some general remarks, we will specialize to the following setting. Given a bijection f of the interval [0,1), define x~y if and only if some (forward, backward, or neutral) iterate of f sends x to y.
We will discuss classical work of Poincare, Denjoy, and Glimm-Effros to the effect that the resulting equivalence relation is either very good or very bad. We will also examine some concrete examples illustrating how hard it can be to decide which alternative holds for a specific f. Along the way, we will catch glimpses into descriptive set theory (defining what "nice" means), dynamics (examining limit sets of orbits), and measure theory (a sneaky way to
show nice uniformizations do not exist).

My interest in this topic arose from joint work with Eugen Ionascu on a question from wavelet theory. If time permits, we will briefly discuss the connection.

Special Seminar in Analysis
2:00-4:00p.m., Ga Tech, Rm 255
Speaker: Jason Metcalfe
Title: Nonlinear wave equations and Strichartz inequalities.

Wavelet Analysis
2:00p.m., Room 326
Speaker: Haipeng Liu, University of Georgia
Title of talk: A Constructive Algorithm for Polynomial Matrix Extension, continued

WEDNESDAY, March 17, 2004

Berkovich Spaces Seminar
11:00a.m., Room 410
Organizers: Matthew Baker and Robert Rumely

Algebra
2:30p.m., Room 410
Speaker:
Jon Kujawa, University of Georgia
Title of talk:
Crystals and the Representation Theory of the Supergroup GL(m|n)
Abstract:
We will discuss the modular representation theory of the supergroup Gl(m|n). In particular, we will see that important aspects of the representation theory are controlled by the combinatorics of an explicitly described crystal graph.

Algebra Geometry Seminar
2:30p.m., Room 303
Speaker:
Klaus Hulek, University of Hanover
Title of talk:
Small algebraic sets
Abstract:
It is well known that an irreducible non-degenerate variety of minimal degree is either a plane, a quadric hypersurface, a rational normal scroll, or a cone over a Veronese surface. There is a similar classification, due to Xambo, for reducible, equi-dimensional algebraic sets connected in codimension 1. We call a (reduced) algebraic set X small if for any plane L which intersects X in a finite scheme the degree of the intersection of X and L is at most dim L + 1. This is the right generalization of minimal degree varieties in the non equi-dimensional case. We prove that X is small if and only if it is 2-regular and show that every such set can be built from a sequence of linearly joined minimal degree varieties. This is a report on joint work with
D. Eisenbud, M. Green and S. Popescu.

Faculty and Graduate Social
3:00p.m., Room 409
Coffee, Cookies, Tea

Numerical Analysis
3:30p.m., Room 303
No Meeting this week

Number Theory
3:45p.m., Room 304
Speaker: Eric Brussel, Emory University
Title of talk: Bilinear forms and Galois cohomology
Abstract: Suppose $k$ is a field that contains all roots of unity. The famous theorem of Merkurjev-Suslin states that the Galois cohomology group $H^2(G_k,\mu)$ is generated by cup products. We discuss the philosophy behind this theorem and its generalization, the Bloch-Kato conjecture, recently claimed by Voevodsky. We show how to use the Merkurjev-Suslin Theorem to compute a ``behavioral ideal'' for the Brauer group of $k$. We apply similar ideas to the case where $k$ is a generalized local field, generalizing part of Hasse's classical theory of the Brauer group of a local field.


THURSDAY, March 18, 2004

VIGRE - Cardiac Physiology
2:30p.m., Room 304
Speakers:
TBA
Activity:
TBA

VIGRE - Contact Topology
2:00p.m., Room 410
Organizer: Gordana Matic, University of Georgia

Faculty and Graduate Social
3:00p.m., Room 409
Coffee, Tea, Cookies

Colloquium
3:30p.m., Room 328
Speaker:
Dr. Thomas Grandine, from the Math. Group of the Boeing Company, Seattle, Washington
Title of talk:
Geometry Generation for Analysis and Optimization
Abstract:
Preliminary design requires timely engineering analysis of large numbers of designs. This leads to the requirement that geometric models be created both rapidly and automatically. Moreover, geometric models so created must necessarily be part of a continuous parametric family of designs, all of which satisfy a large number of shape and spatial integration requirements for the intended vehicle family. Although commercially available geometry systems have some of the attributes needed to satisfy
geometry generation requirements for preliminary design (e.g. many are parametric), none are ideally suited for this application. For many years (decades, in fact), we have been pursuing the holy grail of advanced geometry generation tools which satisfy both the stated and unstated requirements imposed by preliminary design of advanced vehicles. This talk will describe the status of our quest as embodied by the Boeing General Geometry Generator (GGG), our current best effort which has enjoyed some success in design space exploration and optimization.

Student Number Theory

3:30p.m., Room 303
Speaker: TBA
Title of talk: TBA

FRIDAY, March 19, 2004

CATS
1:25-2:15pm, Room 306
Speaker: TBA
Title of talk: TBA

Electrodynamics Seminar
2:30p.m., Room 322
Speaker: Cal Burgoyne, University of Georgia
Title of talk: Introduction to the quantization of the Maxwell and Dirac

Geometry
2:30p.m., Room 326
Speaker: John Sullivan, TU Berlin.
Title of talk: Ropelength, cloth, and clasps, or something else entirely