The reduced-denominator multiplicity conjecture for hard squares on cylinders

Let PmP_m be the path graph on mm vertices and CnC_n the cycle graph on nn vertices. Define

fn(t)=m=0Z(Pm×Cn)tm,f_n(t)=\sum_{m=0}^{\infty} Z(P_m\times C_n)t^m,

where Z(G)=1χ(Ind(G))Z(G)=1-\chi(\operatorname{Ind}(G)) is the Witten index associated with the independence complex of GG. Write each rational function fn(t)f_n(t) after cancelling common factors between its numerator and denominator. Reduced-denominator multiplicity conjecture. After this reduction, f4n+2(t)f_{4n+2}(t) has a denominator with no multiple zeroes, and f4n(t)f_{4n}(t) has a denominator whose zeroes have multiplicity at most two. Consequently, for every fixed nn, the sequence am=Z(Pm×C4n+2)a_m=Z(P_m\times C_{4n+2}) is periodic, while the sequence am=Z(Pm×C4n)a_m=Z(P_m\times C_{4n}) has linear growth. This is a proposed refinement of the established rationality and root-of-unity result for the generating functions; the asserted denominator multiplicity and resulting periodicity or linear-growth conclusions remain open in the source.

Sources & referencesView supporting material

Primary source

Michal Adamaszek, “Hard squares on cylinders revisited”, arXiv:1202.1655 (2012).

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.