PuSH - Publikationsserver des Helmholtz Zentrums München

Becker, S. ; Klein, M.* ; Neitz, A.* ; Parascandolo, G.* ; Kilbertus, N.

Predicting Ordinary Differential Equations with Transformers.

In: (Proceedings of Machine Learning Research). 2023. 1978-2002 (Proceedings of Machine Learning Research ; 202)
We develop a transformer-based sequence-to-sequence model that recovers scalar ordinary differential equations (ODEs) in symbolic form from irregularly sampled and noisy observations of a single solution trajectory. We demonstrate in extensive empirical evaluations that our model performs better or on par with existing methods in terms of accurate recovery across various settings. Moreover, our method is efficiently scalable: after one-time pretraining on a large set of ODEs, we can infer the governing law of a new observed solution in a few forward passes of the model.
Weitere Metriken?
[➜Einloggen]
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Konferenzbeitrag
Korrespondenzautor
Konferenztitel Proceedings of Machine Learning Research
Quellenangaben Band: 202, Heft: , Seiten: 1978-2002 Artikelnummer: , Supplement: ,
Nichtpatentliteratur Publikationen
Institut(e) Helmholtz Artifical Intelligence Cooperation Unit (HAICU)
Institute of Computational Biology (ICB)