fig1

Phthalonitrile melting point prediction enabled by multi-fidelity learning

Figure 1. General implementation route of this work. Limited experimental data are expanded with simulated data, which are utilized to build multi-fidelity models by either (A) ECMs or (B) multi-fidelity co-training methods. Based on these models, we can analyze the contribution of spacers/substituents to PN melting points. ECMs: Error correction method; PN: phthalonitrile.

Journal of Materials Informatics
ISSN 2770-372X (Online)
Follow Us

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/