Generative Learning for Forecasting the Dynamics of Complex Systems

arXiv:2402.17157v1 Announce Type: new
Abstract: We introduce generative models for accelerating simulations of complex systems through learning and evolving their effective dynamics. In the proposed Generative Learning of Effective Dynamics (G-LED), instances of high dimensional data are down sampled to a lower dimensional manifold that is evolved through an auto-regressive attention mechanism. In turn, Bayesian diffusion models, that map this low-dimensional manifold onto its corresponding high-dimensional space, capture the statistics of the system dynamics. We demonstrate the capabilities and drawbacks …

abstract arxiv attention auto complex systems cs.lg data dynamics forecasting generative generative models instances manifold physics.comp-ph physics.flu-dyn simulations stat.ml systems through type