The trichotomy conjecture for the cohomology of simple modules under the linearisation map

About 5 years old · traced to

Let A≅kQ/IA\cong \Bbbk Q/I be an NCCR of the ring RR satisfying Assumption~, and suppose v0=1v_0=1 for some 0∈Q00\in Q_0. Set X=Mθ(A,v)X=\mathcal{M}_\theta(A,v) for θ∈C+\theta\in C_+. For each vertex i≠0i\neq 0, consider the image Ψ(Si)\Psi(S_i) under the linearisation map and its cohomology sheaves.

Trichotomy conjecture. For every vertex i≠0i\neq 0, there is a unique k(i)∈−1,0k(i)\in\\{-1,0\\} such that

Hk(i)(Ψ(Si))≠0.\mathcal{H}^{k(i)}(\Psi(S_i))\neq 0.

Moreover, k(i)=0k(i)=0 if and only if the locus

θ∈Θ∣θi=0\\{\theta\in\Theta\mid \theta_i=0\\}

is a supporting hyperplane of C+C_+.

This conjecture predicts that the nonzero cohomology of each Ψ(Si)\Psi(S_i) occurs in exactly one of two adjacent degrees, with the degree determined by whether the corresponding coordinate locus supports the chamber C+C_+. The supplied text does not state a resolution or provide evidence establishing the conjecture, so its status remains open.

References

Primary source

Alastair Craw, “Gale duality and the linearisation map for noncommutative crepant resolutions”, arXiv:2109.09565 (2021).

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.