Pardey–Rautenbach conjecture on almost colour-balanced perfect matchings

About 2 years old · traced to

Let nn and kk be positive integers, let K2knK_{2kn} be the complete graph on 2kn2kn vertices, and let [k]={1,2,…,k}[k]=\{1,2,\dotsc,k\}. A colouring c ⁣:E(K2kn)→[k]c\colon E(K_{2kn})\to [k] is colour-balanced when every colour is used on the same number of edges. For a perfect matching MM of K2knK_{2kn}, define

f(M)≔∑i=1k∣∣c−1(i)∩M∣−n∣.f(M)\coloneqq \sum_{i=1}^k\left|\left|c^{-1}(i)\cap M\right|-n\right|.

Pardey–Rautenbach conjecture. If c ⁣:E(K2kn)→[k]c\colon E(K_{2kn})\to [k] is colour-balanced, then there is a perfect matching MM of K2knK_{2kn} such that

f(M)≤O(k2).f(M)\leq \mathcal{O}(k^2).

The conjecture seeks a bound depending only on the number of colours, rather than on the number of vertices. Pardey and Rautenbach proposed it after showing that an exactly colour-balanced perfect matching need not exist for three colours; the paper improves the previous bound, which depended on nn, to an explicit bound of 4k24^{k^2}, but does not establish the conjectured quadratic dependence.

References

Primary source

Lawrence Hollom, “A uniform bound on almost colour-balanced perfect matchings in colour-balanced cliques”, arXiv:2410.07993 (2024).

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.