English
 
Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  Stabilized Neural Differential Equations for Learning Dynamics with Explicit Constraints

White, A., Kilbertus, N., Gelbrecht, M., Boers, N. (2024): Stabilized Neural Differential Equations for Learning Dynamics with Explicit Constraints. - In: Oh, A., Naumann, T., Globerson, A., Saenko, K., Hardt, M., Levine, S. (Eds.), Advances in Neural Information Processing Systems 36 (NeurIPS 2023), San Diego : Neural Information Processing Systems, 12929-12950.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://publications.pik-potsdam.de/pubman/item/item_30047 Version Permalink: https://publications.pik-potsdam.de/pubman/item/item_30047_1
Genre: Book Chapter

Files

show Files
hide Files
:
White_2306.09739.pdf (Any fulltext), 7MB
 
File Permalink:
-
Name:
White_2306.09739.pdf
Description:
-
Visibility:
Private
MIME-Type / Checksum:
application/pdf
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-

Creators

show
hide
 Creators:
White, Alistair1, Author              
Kilbertus, Niki2, Author
Gelbrecht, Maximilian1, Author              
Boers, Niklas1, Author              
Affiliations:
1Potsdam Institute for Climate Impact Research, ou_persistent13              
2External Organizations, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Many successful methods to learn dynamical systems from data have recently been introduced. However, ensuring that the inferred dynamics preserve known constraints, such as conservation laws or restrictions on the allowed system states, remains challenging. We propose stabilized neural differential equations (SNDEs), a method to enforce arbitrary manifold constraints for neural differential equations. Our approach is based on a stabilization term that, when added to the original dynamics, renders the constraint manifold provably asymptotically stable. Due to its simplicity, our method is compatible with all common neural differential equation (NDE) models and broadly applicable. In extensive empirical evaluations, we demonstrate that SNDEs outperform existing methods while broadening the types of constraints that can be incorporated into NDE training.

Details

show
hide
Language(s): eng - English
 Dates: 2024-08-012024-08-01
 Publication Status: Finally published
 Pages: 22
 Publishing info: -
 Table of Contents: -
 Rev. Type: No review
 Identifiers: PIKDOMAIN: RD4 - Complexity Science
Organisational keyword: RD4 - Complexity Science
Organisational keyword: FutureLab - Artificial Intelligence in the Anthropocene
Research topic keyword: Nonlinear Dynamics
Model / method: Machine Learning
MDB-ID: pending
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Advances in Neural Information Processing Systems 36 (NeurIPS 2023)
Source Genre: Book
 Creator(s):
Oh, A.1, Editor
Naumann, T.1, Editor
Globerson, A.1, Editor
Saenko, K.1, Editor
Hardt, M.1, Editor
Levine, S.1, Editor
Affiliations:
1 External Organizations, ou_persistent22            
Publ. Info: San Diego : Neural Information Processing Systems
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 12929 - 12950 Identifier: ISBN: 9781713899921