Preprint: mean-field accuracy on large random graphs

The effects of initial conditions on the accuracy of mean-field approximations of Markov processes on large random graphs, joint with Dániel Keliger, is on arXiv: 2506.12872.

The accuracy of a mean-field description turns out to depend on the initial condition, not on network density alone: for generic initial conditions the error is of order $d^{-1/2}$, improving to $1/d + N^{-1/2}$ when the initial state is fairly homogeneous. Read more