Improving the noise estimation of latent neural stochastic differential equations
Chaos: An Interdisciplinary Journal of Nonlinear Science · 01.06.2025 · 35 (6)
DOI: 10.1063/5.0257224 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chaos: An Interdisciplinary Journal of Nonlinear Science, online erschienen am 24.06.2025. Verfasst von L Heck und 3 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Computing and Information Sciences, Radboud University, Nijmegen, The Netherlands. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Volkswagen Foundation, HORIZON EUROPE European Research Council, Land Nordrhein-Westfalen.
Abstract (Originalsprache)
Latent neural stochastic differential equations (SDEs) have recently emerged as a promising approach for learning generative models from stochastic time series data. However, they systematically underestimate the noise level inherent in such data, limiting their ability to capture stochastic dynamics accurately. We investigate this underestimation in detail and propose a straightforward solution; by including an explicit additional noise regularization in the loss function, we are able to learn a model that accurately captures the diffusion component of the data. We demonstrate our results on a conceptual model system that highlights the improved latent neural SDE’s capability to model stochastic bistable dynamics.
Autor:innen und Institutionen
- Institute for Computing and Information Sciences, Radboud University, Nijmegen, The Netherlands.
- Earth System Modelling, School of Engineering and Design, TU Munich, Munich, Germany.
- Computational Network Science, Department of Computer Science, RWTH Aachen University, Aachen, Germany.
ORCID-Profile: L Heck, M Gelbrecht, M T Schaub, N Boers
Förderung
- Volkswagen Foundation
- HORIZON EUROPE European Research Council
- Land Nordrhein-Westfalen
Zitieren
APA
Heck, L., Gelbrecht, M., Schaub, M. T., & Boers, N. (2025). Improving the noise estimation of latent neural stochastic differential equations. Chaos: An Interdisciplinary Journal of Nonlinear Science, 35(6). https://doi.org/10.1063/5.0257224
Vancouver
Heck L, Gelbrecht M, Schaub MT, Boers N. Improving the noise estimation of latent neural stochastic differential equations. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2025;35(6). doi:10.1063/5.0257224
BibTeX
@article{Heck2025,
author = {Heck, L and Gelbrecht, M and Schaub, M T and Boers, N},
title = {Improving the noise estimation of latent neural stochastic differential equations},
journal = {Chaos: An Interdisciplinary Journal of Nonlinear Science},
year = {2025},
volume = {35},
number = {6},
doi = {10.1063/5.0257224}
}
EUVIMED