Conjecture on inclusion divergence for sparse random geometric graphs

About 5 years old · traced to

Let G(n,p)\mathcal{G}(n,p) denote the Erdős–Rényi random graph, let G(n,p,d)\mathcal{G}(n,p,d) denote the random geometric graph in dimension dd, and let p0p_0 be the associated comparison parameter. Suppose that

p∼c0/nfor some c0>0,p\sim c_0/n \quad\text{for some } c_0>0,

and

d∼c1nlog⁡2nfor some c1>0.d\sim c_1n\log^2 n \quad\text{for some } c_1>0.

Inclusion-divergence conjecture. Under these assumptions,

I-Div⁡(G(n,p,d) ∥ G(n,p0))→0.\operatorname{I-Div}\big(\mathcal{G}(n,p,d)\,\|\,\mathcal{G}(n,p_0)\big)\to 0.

This conjecture concerns the boundary at which the random geometric model becomes equivalent to its Erdős–Rényi comparison in inclusion divergence. The supplied text does not state whether the conjecture has been resolved.

References

Primary source

Elliot Paquette and Andrew Vander Werf, “Random geometric graphs and the spherical Wishart matrix”, arXiv:2110.10785 (2021).

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.