Finding separatrices of dynamical flows with Deep Koopman Eigenfunctions
This work addresses the problem of analyzing complex dynamical systems, such as neural circuits, for researchers in neuroscience and related fields, though it appears incremental as it builds on existing Koopman and deep learning methods.
The authors tackled the challenge of characterizing separatrices in high-dimensional dynamical systems by introducing a numerical framework that combines Koopman Theory with Deep Neural Networks to approximate Koopman Eigenfunctions, enabling efficient location of separatrices and design of optimal perturbations for shifting systems across boundaries.
Many natural systems, including neural circuits involved in decision making, are modeled as high-dimensional dynamical systems with multiple stable states. While existing analytical tools primarily describe behavior near stable equilibria, characterizing separatrices--the manifolds that delineate boundaries between different basins of attraction--remains challenging, particularly in high-dimensional settings. Here, we introduce a numerical framework leveraging Koopman Theory combined with Deep Neural Networks to effectively characterize separatrices. Specifically, we approximate Koopman Eigenfunctions (KEFs) associated with real positive eigenvalues, which vanish precisely at the separatrices. Utilizing these scalar KEFs, optimization methods efficiently locate separatrices even in complex systems. We demonstrate our approach on synthetic benchmarks, ecological network models, and high-dimensional recurrent neural networks trained on either neuroscience-inspired tasks or fit to real neural data. Moreover, we illustrate the practical utility of our method by designing optimal perturbations that can shift systems across separatrices, enabling predictions relevant to optogenetic stimulation experiments in neuroscience.