Paul Kainen
DEPARTMENT OF MATHEMATICS
email: kainenp@gusun.georgetown.edu
ICOS 324
1. What is a neural network?
* The
basic concept and some of its history
* Behavior
of the elementary units
* Feedforward
vs. fully interconnected topologies
* Connectionism
and computation
* Functional
approximation and representation
* Finding
the weights
2. Analysis and neural networks.
*
Polynomial approximation
* Weierstrass
and Chebyshev contributions
* Normed
linear spaces
* Dense
subsets as universal approximation and representation
* Finding
the weights
3. Geometry and neural networks.
*
Affine
geometry and uniqueness of parameterization
* Geometry
of balls, cubes and hyperoctahedra - implications of approximation
* Geometry
of the unit ball and best approximation
* Convexity
4. Statistics and neural networks.
*
Probablistic
neurons
* Adding
noise to the weights
* Adding
noise to the activation function
* Pattern
recognition
* Quasi-orthogonality
* Signal
Processing
* Quantum
computing
5. Graph theory and neural networks.
*
Implementation
constraints and bounded vertex degree
* Graph
invariants relevant to neural network design
* Random
graphs and their properties
* Topology
of graphs and implications for implementation
There will be no text for the course which will be based on the proposer's notes, supplemented with xeroxed copies of portions of various books and papers. This is intended to be a research seminar for undergraduates. It is assumed that students have "mathematical maturity" but there are no other prerequisites.
Neural Networks have arisen independently
in computer science, cognitive science and mathematics. Much of the
recent work in quantum computing can be rephrased in terms of neural networks
and amounts to yet another independent rediscovery, this time by physicists.
Of course, neuroscience, philosophy
and psychology also have studied the concept.
Neural networks have a pedigree in mathematics which includes Hilbert, Arnold and Kolmogorov, and they turn out to involve and connect quite a few rather disparate areas - notably, analysis, geometry, graph theory and statistics. Further, neural networks provide a new perspective on nonlinear approximation and will likely have applications to nonlinear optimization.
Therefore, in addition to the rising practical interest in this topic, neural networks are an appropriate subject for theoretical mathematics. Conversely, those from other disciplines who wish to investigate neural networks will be well served by having knowledge of the foundations. Finally, it is to be hoped that the "glamour" of the subject will motivate a new group of people to study math.
For further information, please contact Paul Kainen, Department of Mathematics, Georgetown University. Phone # 202-687-2703