Northeastern Section of the Mathematical Association of America
Fall 2004 Meeting - WPI
Abstracts of Invited Presentations
Quantum mechanics and combinatorial designs
-
P. K. Aravind
Recent proofs of Bell's theorem in the foundations of quantum mechanics have
led into some unexpected areas of discrete mathematics such as regular
polytopes, projective configurations and combinatorial designs. In particular,
this work has led to the emergence of the concept of a "quantum block design"
that is a generalization of the balanced incomplete block design known to
experts in combinatorics and statistical design. The purpose of this talk
is to expose this connection between quantum mechanics and combinatorial
mathematics for the benefit of mathematicians who know next to nothing
about quantum mechanics. This will be done by focusing on a particular
example that illustrates this connection in detail rather than by dwelling
on abstract generalizations. Having done this, I will discuss some open
questions connected with quantum block designs that mathematicians may feel
inclined to address. The possible physical applications of quantum block
designs will also be touched upon.
Some things I should have known when I first started teaching
differential equations (and didn't!)
-
Joseph McKenna
I taught my first course in differential equations thirty
years ago. It was a basic cookbook course and I just followed the
textbook slavishly, often barely grasping what I was teaching. Here, I
will describe some of the things I've learned since I started.
Elliptic curves, the silver bullets of modern mathematics
-
Ezra 'Bud' Brown
Elliptic curves made their first appearance seventeen centuries
ago and are among the most beautiful objects in mathematics --- and the
most useful. This talk will be about elliptic curves and their connections
to such things as:
- Doughnuts and 49/20
- A page from Diophantus
- Magic squares, finite geometries and inflection points
- Congruent numbers and other diophantine problems
- Fermat's Last Theorem
- Factoring Large Primes
- Chocolate-Key Cryptography
A talk on all of these would run for days, so the audience will choose the
topics to be presented.
Alternating Sign Matrices
-
David Bressoud
This will be an overview of what is known and what is conjectured
about Alternating Sign Matrices, a combinatorial structure with ties to
partition theory, representation theory, and statistical mechanics. The
talk will include an overview of the story of the Alternating Sign Matrix
Conjecture, a tale that begins with a Lewis Carroll algorithm for
evaluating determinants and ends with Kuperberg's realization that the
6-vertex model of Izergin and Korepin held the key to the solution.
The CUPM Curriculum Guide 2004
-
David Bressoud
The MAA Committee on the Undergraduate Program in Mathematics (CUPM)
updates its recommendation for the undergraduate curriculum in mathematics
roughly every ten years. The most recent and the most extensive set of
recommendations ever produced by the CUPM was published in February: CUPM
Curriculum Guide 2004. This is the first CUPM curriculum guide to look at
all mathematics courses and the needs of all students taking mathematics
rather than dealing solely with the preparation of majors in the
mathematical sciences. This workshop will explain what can be found in this
guide and how it can be used.
Counting on Determinants
-
Arthur Benjamin
We demonstrate how determinants solve many interesting combinatorial
problems. Determinants count nonintersecting lattice paths, spanning
trees, and permutations with specified descent points. Elegant proofs
of these results are based on the definition of the determinant
and occasionally the principle of inclusion-exclusion. Applications to
Pascal's Triangle, Fibonacci numbers and Catalan numbers will also be
given.
This talk is based on joint work with Naomi Cameron of Occidental
College.
Folding Robot Arms, Proteins, Origami: a Combinatorial Approach
-
Ileana Streinu
Robot arms can be modeled as simple polygonal chains with rigid bars and
rotating joints. Planning non-colliding motions between two configurations
of a robot arm is a notoriously hard problem, for which the currently
known best algorithms run in exponential time. An efficient solution in
dimension 3 could have an impact in understanding apparently unrelated
questions, such as how proteins fold.
In this talk I will present a suprisingly simple combinatorial solution to
the 2-dimensional version. The Carpenter's Rule Problem, "Can every planar
polygonal chain be convexified with non-colliding planar motions?" was
open since the '70. It was answered in the affirmative in the early 2000,
and my contribution - the subject of this talk - was to give an efficient
algorithmic solution. I will present it with a lot of graphical props:
animations, games, even some 3d-graphics. Along the way, we'll use tools
ranging from a 19th century theorem of J. Clerk Maxwell to graph
embeddings, oriented matroids, combinatorial rigidity theory and
visibility computations in Computational Geometry. I will conclude with
some algorithmic insights into origami folding induced by this approach.
Industrial Mathematics: Real Analysis of a Different Kind
-
Arthur Heinricher
Students and faculty at WPI have been working on industrial mathematics
projects for almost 10 years.
The Center for Industrial Mathematics and Statistics at WPI was founded
in 1997 to coordinate these efforts.
This talk will give three snapshots of mathematics at work in business and
industry.
The examples include:
- Differential geometry and optimal design in a steel mill.
- Optimization for cession strategies in automobile insurance.
- Statistical analysis for diagnostic systems in gas stations.
The work described comes from projects completed by graduate and
undergraduate students at WPI as well as work done during the summer in
WPI's Research Experience for Undergraduates in Industrial Mathematics
and Statistics.
Two main messages: First, there are good mathematics problems everywhere,
but the math is often hidden. Second, mathematics students have valuable
skills and can contribute to the solution of real problems of immediate
interest to business and industry.
Spherical Codes, Fullerenes and the Structures of Solutions to the
Problems of Thomson and Tammes
-
Jack Graver
Spherical Codes (maximum families of non overlapping identical spherical caps
on a sphere), Fullerenes (large carbon molecules) and the structures of
solutions to the problems of Thomson (minimize the potential of n unit
charged particles on the sphere) and Tammes (distribute n points on the
sphere to maximize the minimum distance between them) all give rise to
large planar graphs. This talk is about the structures of these graphs
and the relationships between the graphs that arise from different problems.