Equations of some wonderful compactifications (Q416019): Difference between revisions

From MaRDI portal
Changed an Item
ReferenceBot (talk | contribs)
Changed an Item
 
(3 intermediate revisions by 3 users not shown)
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank
Property / OpenAlex ID
 
Property / OpenAlex ID: W2963437356 / rank
 
Normal rank
Property / arXiv ID
 
Property / arXiv ID: 1003.1704 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3941610 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3704017 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Nilpotent subspaces of maximal dimension in semi-simple Lie algebras / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4002278 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Complete collineations revisited / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3360395 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4427516 / rank
 
Normal rank

Latest revision as of 04:47, 5 July 2024

scientific article
Language Label Description Also known as
English
Equations of some wonderful compactifications
scientific article

    Statements

    Equations of some wonderful compactifications (English)
    0 references
    0 references
    0 references
    9 May 2012
    0 references
    The author proves that some wonderful symmetric variety can be defined by linear equations. Let \(G\) be a semisimple adjoint group and let \(G^\sigma\) be the subgroup of fixed points of \(G\) by an involution \(\sigma\). De Concini and Procesi have defined the wonderful compactification \(X\) of \(G/G^\sigma\); it can be described as the closure of the \(G\)-orbit through \(Lie(G^\sigma)\) in the grassmannian of \(Lie(G)\) of the appropriate dimension. In general, the complementary set of \(G/G^\sigma\) in \(X\) is a smooth normal crossing divisor which has exactly \(r\) irreducible components, where \(r\) is the rank of \(X\). Moreover the \(G\)-orbits closures are the intersection of such divisors; in particular there is a unique closed orbit. The author studies the case where the rank of \(X\) is equal to the rank \(r\) of \(G\), i.e. there is maximal (\(\sigma\)-stable) torus over which \(\sigma\) acts as the inverse map. For each \(G\), there is a unique involution with such property; if \(G\) is \(SL_n\), then \(X\) is the variety of complete quadrics. Let \(\omega(x,y,z)=k([x,y],z)\) be the trilinear alternate form associate to the Killing form \(k\). A subspace of \(Lie(G)\) is a nullspace for \(\omega\) if the restriction of \(\omega\) to \(W\) zero; \(W\) is a maximal nullspace if it has maximal dimension between such spaces. Let \(Lie(G)=Lie(G)^\sigma\oplus Lie(G)^{-1}\) be the decomposition in \(\sigma\)-eigenspaces of \(Lie(G)\); then \(Lie(G)^{-1}\) is a nullspace. Moreover the author shows that any Borel subalgebra is a maximal nullspace; hence, by the hypothesis on the rank of \(X\), \(Lie(G)^{-1}\) is a maximal nullspace. Finally, the author shows that \(X\) is isomorphic to the variety of maximal nullspaces by the map \(W\rightarrow W^\perp\) (where \(W^\perp\) is the orthogonal to \(W\) with respect to \(k\)). At the end of this work the author considers two examples of rank 2.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    wonderful compactification
    0 references
    symmetric space
    0 references
    Lie algebra
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references