Topological Quantum Field Theory (291) graduate course, Fall 2004

Topological Quantum Field Theory (291) graduate course, Fall 2004


Time: First lecture Friday 10am, 6438. But how about Monday, Wednesday 4-5.30?

Lecturer: Justin Roberts

"It was twenty years ago today, Vaughan Jones taught us how to play!"

This will be a topics course explaining the theory of "Quantum invariants of three-manifolds" and its relationship with other parts of low-dimensional topology.

The subject began in 1984 with Jones' discovery of his famous polynomial invariant of knots in the three-sphere. This discovery proved to be the tip of the iceberg: several other knot related knot polynomials (HOMFLY, Kauffman, ...) were quickly discovered, and in 1989 Witten explained how this family of invariants should extend naturally to give lots of invariants for three-manifolds. The guiding principle is Topological Quantum Field Theory: an axiomatic characterisation of the class of invariants resembling the Eilenberg-Steenrod homology axioms but of an essentially multiplicative rather than additive character.

Witten's explanation is via physics, including the notorious Feynman path integral technique of quantum field theory. It is still impossible to treat his approach as rigorous, but fortunately the invariants can be studied (in particular, shown to exist!) by many other methods. These span quite a large area of mathematics: chiefly algebra (quantum groups, tensor categories) and differential topology (the rigorous approach to perturbation theory), but there are also elementary combinatorial approaches and a whole lot of interesting diagrammatic methods available.

The course

A provisional outline:
1. Knot polynomials: Jones, HOMFLY, Kauffman, Alexander, and some applications
2. Tensor categories: braids, R-matrices, quantum groups
3. The Chern-Simons path integral and three-manifold invariants
4. Perturbation theory: the Kontsevich integral and configuration space integrals
5. Future directions: Rozansky-Witten invariants; hyperbolic volume conjecture; Gopakumar-Vafa invariants; Khovanov homology


Pre-requisites

It's hard to be very specific. You probably ought to have done the Topology graduate course and know what homology is. Parts of the course (especially at the start, talking about knot invariants) require very little background. Other parts require a bit more abstract algebra, in particular Lie algebras and category theory. I will make sure I explain all the algebra I use though.

Books

A few useful references are as follows:

Quantum Invariants, by Tomotada Ohtsuki (World Scientific)

Knot Theory , by W.B.R. Lickorish (Springer)

Lectures on tensor categories and modular functors, by Bakalov and Kirillov (AMS)

Problems on invariants of knots and 3-manifolds edited by Ohtsuki.

Various lecture notes from a summer school in Grenoble by Vogel, Lescop, etc.

Additional Information

My office hours would be after class on Monday, if the 4pm lecture time goes through.

justin@math.ucsd.edu