Proof of the Knop conjecture (Q2390094)

Let \(G\) be a connected reductive group and \(X\) an irreducible spherical \(G\)-variety. By the weight monoid of \(X\) we mean the submonoid in the weight lattice of \(G\) formed by all highest weights of the \(G\)-module \({\mathbb K}[X]\). In this paper Losev proves a long-standing conjecture due to F.~Knop which states that two smooth affine spherical varieties with the same weight monoid are \(G\)-equivariantly isomorphic. He also provides a criterion allowing to find out whether two affine spherical varieties (not necessarily smooth) are equivariantly isomorphic. Apart of being interesting on their own, these results have some important applications to symplectic geometry. As shown in the last section of Losev's paper, they imply a certain uniqueness result for invariant Kähler structures on a given multiplicity-free compact Hamiltonian manifold with an action of a compact Lie group \(K\). Moreover, these results imply (modulo an unpublished result due to Knop) the Delzant conjecture asserting that any two multiplicity-free compact Hamiltonian \(K\)-manifolds with the same moment polytope and the same principal isotropy subgroup are \(K\)-equivariantly isomorphic.