The strict support-dimension conjecture for Serre relations

Let K=CK=\mathbb{C} and let Π=Π(C,D)\Pi=\Pi(C,D). Define the support of a constructible function f ⁣:nilE(Π,d)Cf\colon \operatorname{nil}_E(\Pi,d)\to\mathbb{C} by

supp(f)={MnilE(Π,d)f(M)0}.\operatorname{supp}(f)=\{M\in\operatorname{nil}_E(\Pi,d)\mid f(M)\not=0\}.

Let M~(Π)\widetilde{{\mathcal M}}(\Pi) be the convolution algebra of constructible functions and let I\mathcal I be the ideal generated by the Serre relations. Strict support-dimension conjecture. For every 0fM~(Π)dI0\not=f\in\widetilde{{\mathcal M}}(\Pi)_d\cap\mathcal I,

dimsupp(f)<dimG(d)qDC(d/D).\dim\operatorname{supp}(f)<\dim G(d)-q_{DC}(d/D).

The preceding dimension estimate gives only the non-strict inequality for arbitrary constructible functions; this conjecture asserts strictness for every nonzero element of the Serre-relation ideal and is the key condition needed to control the quotient by those relations.

Sources & referencesView supporting material

Primary source

Christof Geiß, Bernard Leclerc and Jan Schröer, “Quivers with relations for symmetrizable Cartan matrices IV: Crystal graphs and semicanonical functions”, arXiv:1702.07570 (2018).

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.