Learning dynamically inspired invariant subspaces for Koopman and transfer operator approximation
This work addresses a specific bottleneck in dynamical systems analysis for researchers, though it appears incremental as it builds on existing operator learning methods.
The authors tackled the challenge of efficiently estimating spectra of Koopman and transfer operators from data by machine-learning orthonormal basis functions dynamically tailored to the system, resulting in accurate approximations and nearly invariant subspaces.
Transfer and Koopman operator methods offer a framework for representing complex, nonlinear dynamical systems via linear transformations, enabling a deeper understanding of the underlying dynamics. The spectra of these operators provide important insights into system predictability and emergent behaviour, although efficiently estimating them from data can be challenging. We approach this issue through the lens of general operator and representational learning, in which we approximate these linear operators using efficient finite-dimensional representations. Specifically, we machine-learn orthonormal basis functions that are dynamically tailored to the system. This learned basis provides a particularly accurate approximation of the operator's action as well as a nearly invariant finite-dimensional subspace. We illustrate our approach with examples that showcase the retrieval of spectral properties from the estimated operator, and emphasise the dynamically adaptive quality of the machine-learned basis.