Distinction of the Steinberg representation with respect to a symmetric pair

Seminar
Speaker
Jiandi Zou (Technion)
Date
31/01/2024 - 11:30 - 10:30Add to Calendar 2024-01-31 10:30:00 2024-01-31 11:30:00 Distinction of the Steinberg representation with respect to a symmetric pair Let G be a reductive group over a non-archimedean local field F of odd residual characteristic, let theta be an involution of G over F, and let H be the connected component of the theta-fixed subgroup of G. We are interested in the problem of distinction of the Steinberg representation St of G restricted to H. More precisely, first we give a reasonable upper bound of the dimension of the complex vector space Hom_H(St, C) which was previously known to be finite, and secondly we calculate this dimension for special symmetric pairs (G,H). For instance, the most interesting case for us is when G is a general linear group and H is an orthogonal subgroup of G. Our method follows from the previous results of Broussous-Courtès on Prasad's conjecture. The basic idea is to realize St as the G-space of complex harmonic cochains on the Bruhat-Tits building of G. Thus the problem is somehow reduced to the combinatorial geometry of Bruhat-Tits buildings. This is a joint work with Chuijia Wang. Third floor seminar room and Zoom אוניברסיטת בר-אילן - Department of Mathematics mathoffice@math.biu.ac.il Asia/Jerusalem public
Place
Third floor seminar room and Zoom
Abstract

Let G be a reductive group over a non-archimedean local field F of odd residual characteristic, let theta be an involution of G over F, and let H be the connected component of the theta-fixed subgroup of G. We are interested in the problem of distinction of the Steinberg representation St of G restricted to H. More precisely, first we give a reasonable upper bound of the dimension of the complex vector space

Hom_H(St, C)



which was previously known to be finite, and secondly we calculate this dimension for special symmetric pairs (G,H). For instance, the most interesting case for us is when G is a general linear group and H is an orthogonal subgroup of G.

Our method follows from the previous results of Broussous-Courtès on Prasad's conjecture. The basic idea is to realize St as the G-space of complex harmonic cochains on the Bruhat-Tits building of G. Thus the problem is somehow reduced to the combinatorial geometry of Bruhat-Tits buildings. This is a joint work with Chuijia Wang.

Last Updated Date : 14/02/2024