The exceptional-locus parameterization conjecture for almost large modules

Let AA be a finitely generated basic algebra, module-finite over its prime center ZZ, and let dd be the dimension vector of a large AA-module. Assume that AA is homologically smooth, that ZZ is singular, and that a primitive idempotent eAe\in A satisfies

max{dimC(eW)W a large A-module}=1.\operatorname{max}\left\{\operatorname{dim}_{\mathbb{C}}(eW)\mid W\text{ a large }A\text{-module}\right\}=1.

Assume further that the isoclasses of large AA-modules are parameterized by the smooth locus of MaxZ\operatorname{Max}Z. Exceptional-locus parameterization conjecture. The isoclasses of almost large AA-modules VV satisfying SocV=eSocV\operatorname{Soc}V=e\operatorname{Soc}V are parameterized by the exceptional locus EE of a smooth resolution YMaxZY\to\operatorname{Max}Z. For any fixed path-like set P\mathcal{P} of AA, there is a natural bijection between the irreducible components EiE_i of EE and the distinct subsets PP such that PP is the P\mathcal{P}-annihilator of an almost large module VV with SocV=eSocV\operatorname{Soc}V=e\operatorname{Soc}V, and, if P=PP=P_{\ell} occurs in a maximal chain, the preceding term P1P_{\ell-1} is the P\mathcal{P}-annihilator of a large module. If there exists a sequence of P\mathcal{P}-annihilators

0P1PjP,0\subsetneq P_1\subsetneq\cdots\subsetneq P_j\subsetneq\cdots\subsetneq P_{\ell},

where PjP_j corresponds to EiE_i, then the isoclasses of almost large modules VV with SocV=eSocV\operatorname{Soc}V=e\operatorname{Soc}V and P\mathcal{P}-annihilator PP_{\ell} are parameterized by a codimension-\ell quasi-projective subvariety of EiE_i in YY. The statement relates representation-theoretic degenerations of large modules to the exceptional geometry of a resolution of the singular center; the supplied source gives no evidence that it has been proved or disproved.

Sources & referencesView supporting material

Primary source

Charlie Beil, “The Geometry of Noncommutative Singularity Resolutions”, arXiv:1102.5741 (2011).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.