LS-Galleries, the path model and MV-cycles

We give an interpretation of the path model of a representation \cite{Lit1} of a complex semisimple algebraic group $G$ in terms of the geometry of its affine Grassmannian. In this setting, the paths are replaced by LS--galleries in the affine Coxeter complex associated to the Weyl group of $G$. To explain the connection with geometry, consider a Demazure--Hansen--Bott--Samelson desingularization $\hatΣ(\lam)$ of the closure of an orbit $G(\bc[[t]]).\lam$ in the affine Grassmannian. The homology of $\hatΣ(\lam)$ has a basis given by Białynicki--Birula cell's, which are indexed by the $T$--fixed points in $\hatΣ(\lam)$. Now the points of $\hatΣ(\lam)$ can be identified with galleries of a fixed type in the affine Tits building associated to $G$, and the $T$--fixed points correspond in this language to combinatorial galleries of a fixed type in the affine Coxeter complex. We determine those galleries such that the associated cell has a non-empty intersection with $G(\bc[[t]]).\lam$ (identified with an open subset of $\hatΣ(\lam)$), and we show that the closures of the strata associated to LS-galleries are exactly the MV--cycles \cite{MV}, which form a basis of the representation $V(\lam)$ for the Langland's dual group $G^\vee$.

No reviews yet.

No comments yet.