#machine learning
Jul 2026
The Value of Depth in Message Passing on Sparse Graphs: A Kesten-Stigum Dichotomy
Simulations with an exact belief-propagation baseline on the same trees show that the pairwise rule's error curve is mildly non-monotone in $\ell, so an optimal finite depth exists and BP saturates strictly faster, at an effective per-layer ratio below $\kappa$ that the authors identify.
Aseem Baranwal
· arXiv.org · 0 citations