Automated Construction of Data-Efficient Machine-Learned Force Fields for Polymers
Year: 2026
Type: preprint
Source: ChemRxiv
Institution: University of Luxembourg
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Keywords: machine learning interatomic potentials, polymers, active learning (machine learning), hybrid DFT, data-efficient training, glass transition temperature prediction, density prediction, molecular dynamics, polydimethylsiloxane, hydrogen bonding, polyamide 6, dispersion correction, fine-tuning
Open Access status: closed
OpenAlex ID: W7225918511
