Least-eigenvalue conjecture for set-wise 3-intersecting perfect matchings

From papers

For k11k\geq 11, let A[2k]A_{[2k]}, A[2k2,2]A_{[2k-2,2]}, and A[2k4,2,2]A_{[2k-4,2,2]} be the class adjacency matrices in the perfect matchings association scheme, and define

B^3(2k)=A[2k](k3)(7k10)30(2k4)!!+A[2k2,2]2(k210k+15)15(2k4)!!+A[2k4,2,2]2(k5)5(2k6)!!.\hat{B}_{3}(2k) = A_{[2k]}\frac{(k-3)(7k-10)}{30(2k-4)!!}+A_{[2k-2,2]}\frac{-2(k^{2}-10k+15)}{15(2k-4)!!}+A_{[2k-4,2,2]}\frac{2(k-5)}{5(2k-6)!!}.

Least-eigenvalue conjecture. The row sum and least eigenvalue of B^3(2k)\hat{B}_{3}(2k) are, respectively,

(2k1)(2k3)(2k5)151\frac{(2k-1)(2k-3)(2k-5)}{15}-1

and 1-1. Furthermore, the only modules with eigenvalue equal to 1-1 are [2k2,2][2k-2,2], [2k4,4][2k-4,4], and [2k6,6][2k-6,6].

This conjecture is the spectral ingredient intended to establish the ratio bound with equality for set-wise 33-intersecting perfect matchings. The paper verifies the required eigenvalue bounds for all modules except [2k6,4,2][2k-6,4,2] and [2k6,2,2,2][2k-6,2,2,2], so the remaining cases are open in the stated text.

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

Mahsa N. Shirazi, “An extension of the Erdős-Ko-Rado theorem to set-wise 2-intersecting families of perfect matchings”, arXiv:2110.02175 (2021).

Solutions 0

No solutions have been posted yet.