A reconstruction theorem for almost-commutative spectral triples (Q452229): Difference between revisions

The Connes Reconstruction Theorem says that if \(({\mathcal A}, {\mathcal H}, D)\) is an orientable commutative spectral triple of metric dimension \(p\), then there exists a smooth compact oriented \(p\)-dimensional Riemannian manifold \(X\) such that \({\mathcal A}=C^{\infty}(X)\). By an almost-commutative spectral triple (in the sense of Connes) one understands a Cartesian product of canonical spectral triples of a compact spin manifold (a commutative spectral triple) with a finite spectral triple; it was conjectured by Connes that there exists an analog of the Reconstruction Theorem for the almost-commutative spectral triples. The paper under review is a confirmation of this conjecture. Namely, the author introduces a generalization of almost-commutative spectral triples by which he understands a spectral triple of the form \((C^{\infty}(X,A), L^2(X,H), D)\), where \(X\) is a compact oriented Riemannian manifold, \(H\) a self-adjoint Clifford module bundle, \(A\) a real unital \(\ast\)-algebra subbundle of \(H\) and \(D\) a symmetric Dirac-type operator on \(H\). The main result (a reconstruction theorem) says that for any almost-commutative spectral triple of the above type there exists a compact oriented Riemannian manifold \(X\) such that algebra \(C^{\infty}(X)\) realizes the triple. A corrigendum to the printed version of paper is posted at an updated version of the preprint at \url{arxiv:1101.5908}.