PDE-Driven Spatiotemporal Disentanglement - Archive ouverte HAL Access content directly
Conference Papers Year : 2021

PDE-Driven Spatiotemporal Disentanglement

Abstract

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.
Fichier principal
Vignette du fichier
var_sep.pdf (2.25 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02911067 , version 1 (03-08-2020)
hal-02911067 , version 2 (05-10-2020)
hal-02911067 , version 3 (17-03-2021)

Licence

Attribution

Identifiers

Cite

Jérémie Donà, Jean-Yves Franceschi, Sylvain Lamprier, Patrick Gallinari. PDE-Driven Spatiotemporal Disentanglement. The Ninth International Conference on Learning Representations, International Conference on Representation Learning, May 2021, Vienne, Austria. ⟨hal-02911067v3⟩

Relations

313 View
135 Download

Altmetric

Share

Gmail Facebook X LinkedIn More