The theory behind the synchronization of network-coupled oscillators is solid and has significantly advanced lately, especially in the mean-field problem. However, real systems have complex interactions that difficult the analytical treatment. Most existing results rely on black-boxes that hinder interpretation and require global information, while available data is usually incomplete. In this seminar, I will overview some results of my PhD thesis motivated by these limitations, including methods to predict the non-linear amplification of noise from network weights to the synchronization onset and the role of dyadic and higher-order interactions in synchrony optimization, a geometric unfolding of the synchronized state which explains the emergence of synchronization bombs, and algebraic shortcuts to find critical points using rank-reduced network models. These tools also unveil some mechanistic explanations that were hidden behind the prevalent assumptions of complete information.
Presential seminar, with parallel Zoom stream:
Lucas Lacasa Contact form