Спасибо, ясно.
Ну, например, root lattices

или

?
Есть такой коммент:
"if it's the root lattices

or

then I think you'll get "the same approximation" in the sense that
there will be an asymptotic expansion in inverse powers of d whose
leading term will come from walks that return in only 2 steps, i.e.
probability

where

is the number of neighbors
(a.k.a. degree) of each vertex on the graph."