Wilson's conjecture on non-trivially unstable Rose Window graphs

From papers

Let nn be an integer with n3n\geq 3, and let a,rZna,r\in\mathbb{Z}_n with a,r0a,r\ne 0. The Rose Window graph Rn(a,r)R_n(a,r) has vertex set {ui,vi:iZn}\{u_i,v_i:i\in\mathbb{Z}_n\} and edges

{ui,ui+1},{vi,vi+r},{ui,vi},{ui+a,vi},iZn,\{u_i,u_{i+1}\},\quad \{v_i,v_{i+r}\},\quad \{u_i,v_i\},\quad \{u_{i+a},v_i\},\qquad i\in\mathbb{Z}_n,

where subscripts are computed modulo nn. A graph is non-trivially unstable if it is unstable with respect to its canonical double cover and is neither bipartite nor has two vertices with the same neighborhood. The nine families W1--W9 are the families of unstable Rose Window graphs listed in Table 1 of the source.

Wilson's conjecture. Every non-trivially unstable Rose Window graph is isomorphic to a graph in one of the families W1--W9.

Wilson checked this conjecture computationally for Rose Window graphs on at most 200200 vertices. The conjecture asserts that the nine listed families exhaust all non-trivially unstable Rose Window graphs; the source gives no resolution beyond this computational verification.

Progress summary

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

Sources & referencesView supporting material

Primary source

Milad Ahanjideh, István Kovács and Klavdija Kutnar, “Stability of Rose Window graphs”, arXiv:2306.01619 (2024).

Solutions 0

No solutions have been posted yet.