fig8

Multi-activity derivative automated labeling and margin-sampling active learning for efficient determination of phase boundaries

Figure 8. Final Sampling distributions and ML-reconstructed phase diagrams obtained using MLP combined with MS for (A) Pd-Pt and (B) Au-Ag-Ge. The left panels show the labeled samples, where triangles represent the uniformly distributed initial samples and circles represent the subsequently queried samples; The right panels show the constructed phase regions, whose interfaces represent the ML-predicted phase boundaries. ML: Machine learning; MLP: multi-layer perceptron; MS: margin sampling.

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/