Joint embedding predictive architecture for self-supervised pretraining on polymer molecular graphs
Digital Discovery · 2026 · 5 (2) · 819-834
DOI: 10.1039/d5dd00308c ↗Kurz gefasst
Fachartikel in der Zeitschrift Digital Discovery, 2026. Verfasst von Francesco Piccoli und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0). Gefördert von: Firmenich, Rijksdienst voor Ondernemend Nederland.
Abstract (Originalsprache)
We address the challenge of limited labeled data in polymer ML, using a Joint Embedding Predictive Architecture (JEPA) for self-supervised pretraining on polymer graphs. This improves downstream performance when finetuning on small labelled datasets.
Förderung
- Firmenich
- Rijksdienst voor Ondernemend Nederland
Zitieren
APA
Piccoli, F., Vogel, G., & Weber, J. M. (2026). Joint embedding predictive architecture for self-supervised pretraining on polymer molecular graphs. Digital Discovery, 5(2), 819-834. https://doi.org/10.1039/d5dd00308c
Vancouver
Piccoli F, Vogel G, Weber JM. Joint embedding predictive architecture for self-supervised pretraining on polymer molecular graphs. Digital Discovery. 2026;5(2):819-834. doi:10.1039/d5dd00308c
BibTeX
@article{Piccoli2026,
author = {Piccoli, Francesco and Vogel, Gabriel and Weber, Jana M.},
title = {Joint embedding predictive architecture for self-supervised pretraining on polymer molecular graphs},
journal = {Digital Discovery},
year = {2026},
volume = {5},
number = {2},
pages = {819--834},
doi = {10.1039/d5dd00308c}
}
EUVIMED