DFORM: Diffeomorphic vector field alignment for assessing dynamics across learned models

arXiv:2402.09735v1 Announce Type: new
Abstract: Dynamical system models such as Recurrent Neural Networks (RNNs) have become increasingly popular as hypothesis-generating tools in scientific research. Evaluating the dynamics in such networks is key to understanding their learned generative mechanisms. However, comparison of learned dynamics across models is challenging due to their inherent nonlinearity and because a priori there is no enforced equivalence of their coordinate systems. Here, we propose the DFORM (Diffeomorphic vector field alignment for comparing dynamics across learned models) …

abstract alignment arxiv become comparison cs.lg cs.sy dynamics eess.sy generative hypothesis key networks neural networks popular q-bio.nc recurrent neural networks research scientific research tools type understanding vector