Improving the Noise Estimation of Latent Neural Stochastic Differential Equations

作     者：Heck, Linus Gelbrecht, M. Schaub, M. Boers, N. 

作者机构：Institute for Computing and Information Sciences Radboud University Nijmegen Netherlands Earth System Modelling School of Engineering and Design TU Munich Germany Computational Network Science Dept of Computer Science RWTH Aachen University Germany Department of Computer Science RWTH Aachen University Germany Complexity Science Potsdam Institute for Climate Impact Research Potsdam Germany Department of Mathematics and Statistics University of Exeter Exeter United Kingdom 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2024年

核心收录：

主　　题：Stochastic models 

摘      要：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. © 2024, CC BY-NC-ND.