PDE-Driven Spatiotemporal Disentanglement
This addresses the problem of accurate spatiotemporal prediction for applications like video analysis, though it appears incremental by building on existing differential equation approaches.
The paper tackles predicting high-dimensional spatiotemporal phenomena by proposing a novel paradigm based on separation of variables from PDE theory, which learns disentangled spatial and temporal representations to improve prediction accuracy, demonstrating performance against prior state-of-the-art models on physical and synthetic video datasets.
A recent line of work in the machine learning community addresses the problem of predicting high-dimensional spatiotemporal phenomena by leveraging specific tools from the differential equations theory. Following this direction, we propose in this article a novel and general paradigm for this task based on a resolution method for partial differential equations: the separation of variables. This inspiration allows us to introduce a dynamical interpretation of spatiotemporal disentanglement. It induces a principled model based on learning disentangled spatial and temporal representations of a phenomenon to accurately predict future observations. We experimentally demonstrate the performance and broad applicability of our method against prior state-of-the-art models on physical and synthetic video datasets.