Is there a “fermionized”, or partially odd version of the above con- struction?
Statement
Is there a “fermionized”, or partially odd version of the above con- struction? I.e. can we recover Yangians of classical super-algebras like gl(m|n)? Is there a choice of quiver Q′ and potential W ′ such that AQ′ ,W ′ “partially fermionizes” AQ̃,W̃ ? where aQ,\gamma (q) is the Kac polynomial, counting iso-classes of absolutely indecomposable \gamma-dimensional Fq Q-modules. (3) There is an isomorphism of Lie algebras g0ΠQ \sim = n−Qre where Q re is the − quiver obtained by removing all vertices supporting loops, and nQre is one half of the associated Kac–Moody Lie algebra. (4) (Work in progress, with L. Hennecart and S. Schlegel-Mejia): The Lie algebra gΠQ is one half of a generalised Kac–Moody Lie algebra. (1), (3) and (4) suggest that AΠQ is one half of some kind of generalized Yangian. Indeed in finite and affine type, this statement can be made precise. (2) tells us that the Lie algebra/algebra is concentrated entirely in even cohomological degrees. We are motivated by the We will answer this question by taking the word “fermionization” seriously: physicists teach us that particles acquire mass (i.e. become fermionic) by adding “mass terms” or quadratic terms, to superpotentials governing their gauge the- ory. So first we have to recast AΠQ in terms of potentials. A potential W \in CQ/[CQ, CQ]cyc is a linear combination P of cyclic paths. If W = a1 . . . an and a \in Q1 we define \partialW/\partiala = ai =a ai+1 . . . an a1 . . . ai−1 and extend to arbitrary W by linearity. Then define Jac(Q, W ) := CQ/h\partialW/\partiala | a \in Q1 i. An important class of examples come from the tripling construction. We denote by Q̃ the quiver obtained by adding a loop \omegai at P every vertex ofPQ, the doubled quiver. This carries the cubic potential W̃ = ( a\inQ1 [a, a∗ ])( i\inQ0 \omegai ). It is easy to verify \sim P that there is a natural isomorphism Ψ : Jac(Q̃, W̃ ) = ΠQ [\omega] with Ψ−1 (\omega) = i\inQ0 \omegai . Given a quiver Q with potential W we denote by Tr(W ) \in Γ(M(CQ)) the re- sulting function on the stack of CQ-modules. Then as substacks of M(CQ) there are equalities crit(Tr(W )) = M(Jac(Q, W )) = supp(\phiTr(W ) Qvir ) where \phiTr(W ) Qvir is a perverse sheaf on the stack of Jac(Q, W )-modules that turns out to be the
Record
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.