A method for unsupervised learning of coherent spatiotemporal patterns in multiscale data.

Abstract

The unsupervised and principled diagnosis of multiscale data is a fundamental obstacle in modern scientific problems from, for instance, weather and climate prediction, neurology, epidemiology, and turbulence. Multiscale data are characterized by a combination of processes acting along multiple dimensions simultaneously, spatiotemporal scales across orders of magnitude, nonstationarity, and/or invariances such as translation and rotation. Existing methods are not well-suited to multiscale data, usually requiring supervised strategies such as human intervention, extensive tuning, or selection of ideal time periods. We present the multiresolution coherent spatio-temporal scale separation (mrCOSTS), a hierarchical and automated algorithm for the diagnosis of coherent patterns or modes in multiscale data. mrCOSTS is a variant of dynamic mode decomposition which decomposes data into bands of spatial patterns with shared time dynamics, thereby providing a robust method for analyzing multiscale data. It requires no training but instead takes advantage of the hierarchical nature of multiscale systems. We demonstrate mrCOSTS using complex multiscale datasets that are canonically difficult to analyze: 1) climate patterns of sea surface temperature, 2) electrophysiological observations of neural signals of the motor cortex, and 3) horizontal wind in the mountain boundary layer. With mrCOSTS, we trivially retrieve complex dynamics that were previously difficult to resolve while additionally extracting hitherto unknown patterns of activity embedded in the dynamics, allowing for advancing the understanding of these fields of study. This method is an important advancement for addressing the multiscale data which characterize many of the grand challenges in science and engineering.