Bi-FORK: Generative Modeling of High-Dimensional Bifurcating Systems
Bifurcations are ubiquitous in physical systems, from structural buckling to fluid and climate dynamics, yet they remain largely unexplored in deep learning. At a symmetry-breaking bifurcation, a single input admits multiple equally valid solutions, violating the one-to-one assumption underlying most learned physical surrogates. We introduce Bi-FORK, a generative framework for
Reading in Plain English mode — toggle below to switch
Bifurcations are ubiquitous in physical systems, from structural buckling to fluid and climate dynamics, yet they remain largely unexplored in deep learning. At a symmetry-breaking bifurcation, a single input admits multiple equally valid solutions, violating the one-to-one assumption underlying most learned physical surrogates. We introduce Bi-FORK, a generative framework for
This paper introduces a novel generative framework, Bi-FORK, to address the challenging problem of modeling systems with bifurcations, where a single input can lead to multiple valid outputs. This capability is crucial for accurately simulating complex physical phenomena in engineering and scientific domains, moving beyond the limitations of traditional one-to-one mapping models.