A physics-informed autoencoder compresses 3D charge density into a 16x16x16x16 latent representation that, combined with MAGPIE descriptors, predicts bulk modulus, Young's modulus, shear modulus, formation energy, and Debye temperature with R2 values of 0.94, 0.88, 0.87, 0.96, and 0.89 on 6059 DFT-s
and Toher, Cormac and Curtarolo, Stefano and Ceder, Gerbrand and Persson, Kristin A
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
fields
cond-mat.mtrl-sci 4years
2026 4roles
method 1polarities
use method 1representative citing papers
Averaged covalent and ionic bond strengths correlate with and can estimate oxygen vacancy migration barriers across rutile 3d transition-metal dioxides after fitting two parameters to DFT data.
Quantum-chemical bonding descriptors improve machine learning predictions of materials properties and enable symbolic regression to recover intuitive expressions for force constants and thermal conductivity.
CrabNet outperforms MODNet and random forest models when predicting battery electrode properties from composition, with cross-validation and clustering confirming coherent groupings.
citing papers explorer
-
Physics Aware Representation Learning on Electronic Charge Density for Materials Property Prediction
A physics-informed autoencoder compresses 3D charge density into a 16x16x16x16 latent representation that, combined with MAGPIE descriptors, predicts bulk modulus, Young's modulus, shear modulus, formation energy, and Debye temperature with R2 values of 0.94, 0.88, 0.87, 0.96, and 0.89 on 6059 DFT-s
-
Bond-Strength-Based Understanding of Oxygen Vacancy Migration Barriers in Rutile Oxides
Averaged covalent and ionic bond strengths correlate with and can estimate oxygen vacancy migration barriers across rutile 3d transition-metal dioxides after fitting two parameters to DFT data.
-
A critical assessment of bonding descriptors for predicting materials properties
Quantum-chemical bonding descriptors improve machine learning predictions of materials properties and enable symbolic regression to recover intuitive expressions for force constants and thermal conductivity.
-
Machine Learning for Electrode Materials: Property Prediction via Composition
CrabNet outperforms MODNet and random forest models when predicting battery electrode properties from composition, with cross-validation and clustering confirming coherent groupings.