Meetings of the Rutgers Lie Group Seminar in Spring 1995


Tuesday, Jan. 17, 1995, 3:00pm-4:00pm, Hill 250 (special date) Shrawan Kumar: The Picard group of moduli spaces of G-bundles, and representation theory
Friday, Jan. 27, 1995, 11:30am-12:30pm, Hill 425 Friedrich Knop: The image of the moment map from the differentiable point of view Consider a Hamiltonian action of a compact group K on a symplectic manifold X. We are interested in smooth functions which Poisson-commute with all K-invariant smooth functions on X. All pull-back functions from the moment map have this property but there are (slightly) more of them. We explain how this difference is controlled by a certain subgroup W_X of the Weyl group of K. The group W_X is also a reflection group and is analogous to the little Weyl group of a symmetric space. We discuss also its effect on the automorphism group of X.
Friday, Feb. 3, 1995, 11:30am-12:30pm, Hill 425 Siddhartha Sahi: Euclidean Fourier transforms and unitary representations Consider a (degenerate) principal series representation X attached to a parabolic subgroup P=LN of a semisimple Lie group G. We can regard the vectors in the representation as functions on the opposite nilradical, and then via the Euclidean Fourier transform, as tempered distributions on N. If Z is a submodule of X, then the support of the distributions in Z is a P-orbit on N, closely related to the associated variety of Z. It seems that (possible) unitary structures on Z can often be described rather simply in terms of measures and distributions on P-orbits in N. The talk will discuss applications of this philosophy to the case where N is abelian.
Friday, Feb. 10, 1995, 11:30am-12:30pm, Hill 425 Roger Zierau: Maximal compact subvarieties in flag domains There are many ways to realize representations as sections of a vector bundle over a symmetric space. A Penrose transform often does the trick for representations associated to open orbits in a complex flag variety. The general construction uses a double fibration (between the orbit D and a space of maximal compact subvarieties M(D)) along with some standard geometric tricks. When the representations are highest weight representations M(D)=G/K and the general procedure works very well and is understood. Otherwise one needs to know more about the space M(D). I will discuss some recent joint work with Joe Wolf on the structure of M(D). I will also describe how this leads to a realization of the representation in a completely holomorphic setting.
Friday, Feb. 17, 1995, 11:30am-12:30pm, Hill 425 Kamal Khuri-Makdisi: Fourier coefficients of forms of half-integral weight Let f(z) = \sum_{n > 0} a_n exp(2\pi inz) be a cusp form of half-integral weight m + ½. If n = q^2 r with r square-free, then by Shimura's correspondence the n-th Fourier coefficient a_n can be expressed in terms of a_r and the Fourier coefficients of a form g of integral weight 2m. As for a_r, a theorem of Waldspurger says that its square is more or less the central value of the L-function of g twisted by the quadratic character corresponding to Q(r^½). This talk will describe a method, due to Shimura, to get explicit forms of Waldspurger's relation. The method works well over totally real number fields (even for nonholomorphic forms), and should generalize straightforwardly to forms over an arbitrary number field.
Friday, Feb. 24, 1995, 11:30am-12:30pm, Hill 425 Ranee Brylinski: Geometric quantization and minimal unitary representations In this talk I will give an overview of joint work with Bert Kostant on the construction, following the precepts of geometric quantization, of a unitary representation attached to the minimal nilpotent orbit O. The representations we obtain are natural generalizations of the Fock space model of the metaplectic representation. One feature of our methods is that exceptional Lie groups are treated in exactly the same manner as the classical ones. Roughly speaking, the main point is that, using the geometry and invariant theory of the complexification of O, we find new realizations of a finite-dimensional complex simple Lie algebra g in terms of p's and q's. Then we quantize these symbols to realize g as as a Lie algebra of pseudo-differential operators. This leads to many results on harmonic analysis for the ensuing unitary representation of the underlying real group.
Thursday, March 2, 1995, 3:00pm-4:00pm, Hill 705 Jean-Luc Brylinski: A dual to the McKay correspondence Finite subgroups of SU(2) are associated to Dynkin diagrams of type A, D, E. The McKay correspondence gives a bijection between irreducible representations of the finite group and vertices of the extended Dynkin diagram. The edges give the operation of tensoring with the 2-dimensional representation. The dual correspondence we propose is a bijection between non-trivial conjugacy classes of the finite group and vertices of the diagram.The ends of the diagram give 3 distinguished conjugacy classes, and edges correspond to taking a product with an element in one of these classes. This correspondence is based on the topology of the resolution of singularities of the Kleinian singularity.
Friday, March 3, 1995, 11:30am-12:30pm, Hill 425 Sam Evens: Characteristic cycles for the loop Graßmannian We explain how to compute characteristic cycles of D-modules in several settings relevant to representation theory. The loop Graßmannian is a homogeneous space for a loop group LG with the property that its equivariant D-modules correspond to finite dimensional representations of the Langlands dual of G. In this setting, we compute the characteristic cycles. If time permits, we will discuss some examples related to Springer representations and affine Hecke algebra representations. This talk is based on joint work with Ivan Mirkovic.
Friday, March 10, 1995, 10:30am-11:30am, Hill 425 Yael Karshon: Hamiltonian torus actions We will define a Hamiltonian torus action on a symplectic manifold M. The best understood examples are smooth compact toric varieties. These are in one to one correspondence with a set of polytopes in R^n where n=½dim M. The symplectic quotients of these spaces are single points. Compact spaces whose symplectic quotients have dimension 2 are currently studied; those with dim M = 4 are completely understood. In the "generic" case, the action extends to form a toric variety (where the word "generic" is used in a somewhat loose way).
Friday, March 10, 1995, 12:00pm-1:00pm, Hill 425 Wolfgang Soergel: n-cohomology of limits of discrete series We drag with heavy machinery this problem into the realm of combinatorics of Weyl groups. Some known cases can be recovered this way, and some up to now unknown cases can still be treated in a satisfactory way.
Friday, March 17, 1995, 11:30am-12:30pm, Hill 425 Spring break. No meeting.
Friday, March 24, 1995, 10:30am-11:30am, Hill 425 Peter Heinzner: Equivariant holomorphic vector bundles The "Oka Principle" can be vaguely stated as follows On a reduced Stein space X, problems which can be cohomologically formulated have only topological obstructions. In other words, such problems are holomorphically solvable if and only if they are continuously solvable. For example, Grauert has proved that the categories of topological or complex analytic principal bundles over a Stein space X with a complex Lie group as structure group are the same. In this talk we explain the equivariant version of Grauert's Oka Principle for a compact Lie group of holomorphic transformations on X. Theorem. (Join work with Frank Kutzschebauch) Let K be a compact Lie group of holomorphic transformations of a reduced Stein space X. Then every topological complex K-vector bundle over X is K-isomorphic to a holomorphic K-vector bundle over X. Moreover, two holomorphic K-vector bundles over X are holomorphically K-isomorphic if and only if they are topologically K-isomorphic.
Friday, March 24, 1995, 12:00pm-1:00pm, Hill 425 Joseph Wolf: Flag domains and realization of representations by the double fibration construction The representation theory of semisimple Lie groups makes some fascinating connections between harmonic analysis, differential geometry and analytic number theory. For example one can study representations by constructing them in a variety of geometric settings. I'll survey some of those constructions and describe some recent progress. Level: graduate students should be able to follow most of the talk.
Friday, March 31, 1995, 11:30am-12:30pm, Hill 425 Soji Kaneyuki: Signatures of roots and a new characterization of causal symmetric spaces A signature function on a root system of a real semisimple Lie algebra g is a {±1}-valued multiplicative character of the root lattice. A causal symmetric space is a symmetric space admitting an invariant smooth cone field in tangent spaces. Simple causal symmetric spaces were classified by G.I. Ol'shansky (1982). (1) We show that every signature function arises from a Z-gradation of g. From this point of view, semisimple symmetric spaces of type K_e, introduced by Oshima-Sekiguchi, break up into two classes: K_eI and K_eII. (2) We give a characterization of simple causal symmetric spaces in terms of K_eI.
Friday, April 7, 1995, 11:30am-12:30pm, Hill 425 Toshiyuki Kobayashi: Discrete decomposability of A_q(l) with respect to reductive subgroups and its applications
Friday, April 14, 1995, 11:30am-12:30pm, Hill 425 Robert Shalla: The universal cover of SU(2,1) and algebraic D-module techniques in representation theory Some recent work of Hecht, Milicic, Schmid, and Wolf uses the theory of algebraic D-modules to study the representations of a semisimple Lie group, G, with finite center. A straightforward generalization of their constructions can also produce useful realizations of the irreducible representations of G when the center is infinite. In this talk G will be the universal cover of SU(2,1) and we shall discuss some techniques of algebraic D-modules in this setting. These techniques give a simple description of the irreducible tempered representations and of the composition series of standard modules. We can also understand, in geometric terms, the "new" phenomena which arise in the infinite center case.
Friday, April 21, 1995, 11:30am-12:30pm, Hill 425 Kailash Misra: Application of crystal basis to affine Lie algebra representations The crystal basis associated with integrable highest weight representations of affine Lie algebras provide useful combinatorial tool to study the corresponding representations. In this talk we will discuss how this tool can be used to show that any integrable highest representation of classical affine Lie algebras can be embedded in the tensor product of level one representations of the corresponding affine Lie algebras.
Friday, April 28, 1995, 11:30am-12:30pm, Hill 425 Leon Ehrenpreis: Differential equations and representation theory