Irregular spanning subgraph with minimum degree bound
Conjecture 1.2 · arXiv:2108.02685
Status partial high confidence
Conjecture 1.2 from arXiv:2108.02685 remains open in full generality. Fox, Luo, and Pham (2022) proved the asymptotic version for d-regular graphs when d = o(n/(log n)^{12}) via probabilistic methods. Ma and Xie (2024) obtained the first n-independent bound for d-regular graphs (within 2d^2 of the conjectured value) and proved the conjecture in strong form for d=3 regular graphs using deterministic local-adjustment techniques. Luzar, Przybylo, and Sotak (2024) independently resolved the cubic case exactly, showing deviation from n/4 is at most 1/2 (up to three exceptions). The exact bound for general graphs with minimum degree delta is still open.
Cited literature (3)
-
Proves the asymptotic version of the conjecture for d-regular graphs when d = o(n/(log n)^{12}): with high probability the irregular random subgraph satisfies m(H,k) = (1+o(1))n/(d+1) for all 0 <= k <= d.
-
Gives the first n-independent bound for d-regular graphs (vertex degree counts within 2d^2 of n/(d+1)) and proves the conjecture in strong form for cubic (d=3) regular graphs via deterministic local-adjustment methods.
-
Resolves the Alon-Wei conjecture for cubic graphs, showing that every cubic graph on n vertices has a spanning subgraph where the number of vertices of each degree deviates from n/4 by at most 1/2, up to three exceptions.
Reviewer notes. The conjecture is closely related to Conjecture 1.1 (the regular-graph version). Most follow-up work addresses the regular case; the minimum-degree generalisation of Conjecture 1.2 has seen partial progress only through the special case d=3 and asymptotic results. No paper found that resolves the full conjecture for arbitrary minimum degree delta.
Context
This is the second of the two central conjectures introduced by the authors, generalising Conjecture 1.1 from regular graphs to arbitrary graphs with minimum degree $\delta$. Both conjectures remain open; the paper proves asymptotic relaxations showing the bound $(1+o(1))\frac{n}{\delta+1}+2$ is achievable when $n$ is large.
Notes. Source is PDF extraction; mathematical notation appears clean for this statement.
Source paper
Irregular Subgraphs
Noga Alon, Fan Wei · 2021-08-06
https://arxiv.org/abs/2108.02685
PDF source