Journal article icon

Journal article

"Hey, that's not an ODE": faster ODE adjoints with 12 lines of code

Abstract:

Neural differential equations may be trained by backpropagating gradients via the adjoint method, which is another differential equation typically solved using an adaptive-step-size numerical differential equation solver. A proposed step is accepted if its error, \emph{relative to some norm}, is sufficiently small; else it is rejected, the step is shrunk, and the process is repeated. Here, we demonstrate that the particular structure of the adjoint equations makes the usual choices of norm (s...

Expand abstract
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Files:
Publication website:
http://proceedings.mlr.press/v139/kidger21a.html

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Mathematical Institute
Oxford college:
St Hilda's College
Role:
Author
More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Mathematical Institute
Role:
Author
ORCID:
0000-0002-9972-2809
Publisher:
Journal of Machine Learning Research Publisher's website
Series:
Proceedings of Machine Learning Research
Series number:
139
Pages:
5443-5452
Publication date:
2021-07-01
Acceptance date:
2021-05-08
Event title:
Thirty-eighth International Conference on Machine Learning (ICML 2021)
Event location:
Virtual event
Event website:
https://icml.cc/Conferences/2021
Event start date:
2021-07-18
Event end date:
2021-07-24
ISSN:
2640-3498
Language:
English
Keywords:
Pubs id:
1133477
Local pid:
pubs:1133477
Deposit date:
2021-02-18

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP