Logarithm-free inducibility bound conjecture

For an nn-vertex graph GG, let XG,kX_{G,k} be the number of edges induced by a uniformly random kk-vertex subset of GG. Define

I(n,k,)=max{Pr(XG,k=):V(G)=n}I(n,k,\ell)=\max\{\Pr(X_{G,k}=\ell):|V(G)|=n\}

and

ind(k,)=limnI(n,k,).\operatorname{ind}(k,\ell)=\lim_{n\to\infty}I(n,k,\ell).

For 0(k2)0\le \ell\le\binom{k}{2}, define

=min\originalleft{,(k2)\aftergroup\originalright}.\ell^*=\min\mathopen{}\mathclose\bgroup\originalleft\{\ell,\binom{k}{2}-\ell\aftergroup\egroup\originalright\}.

Logarithm-free inducibility bound conjecture. For all kk and all 0(k2)0\le\ell\le\binom{k}{2}, we have

ind(k,)=O\originalleft(k/\aftergroup\originalright).\operatorname{ind}(k,\ell)=O\mathopen{}\mathclose\bgroup\originalleft(\sqrt{k/\ell^*}\aftergroup\egroup\originalright).

The conjecture is proposed to remove the logarithmic factor appearing in the paper's preceding bound, whose logarithm arises from an estimate related to the Gotsman–Linial conjecture. Its status is not resolved in the supplied text.

Sources & referencesView supporting material

Primary source

Matthew Kwan, Benny Sudakov and Tuan Tran, “Anticoncentration for subgraph statistics”, arXiv:1807.05202 (2018).

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.