Variance-optimal functions determine isoperimetric extremizers

Let GG be a graph, let XX be a variance-optimal function over GG, and for suitable rr define Sr={a=(a1,,an)V(Gn):i=1nX(ai)r}S_r=\{a=(a_1,\ldots,a_n)\in V(G^n):\sum_{i=1}^n X(a_i)\leq r\}. For a vertex set SV(Gn)S\subseteq V(G^n), let Bd(S)={u:dGn(u,S)d}B_d(S)=\{u:d_{G^n}(u,S)\leq d\}.

Isoperimetric extremizer conjecture. For nn sufficiently large and d,rd,r in appropriate ranges,

Bd(Sr)Bd(S)|B_d(S_r)|\leq |B_d(S')|

for every SV(Gn)S'\subset V(G^n) with SrS|S_r|\leq |S'|.

This conjecture proposes that sets formed from variance-optimal functions determine the extremal isoperimetric behavior of Cartesian powers of graphs. The supplied text gives no resolution or precise ranges for dd and rr.

Sources & referencesView supporting material

Primary source

Matthew Yancey, “Probabilistic and Geometrical Applications to Graph Theory”, arXiv:1705.09725 (2017).

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.