Active set algorithms enable automated data-driven sparse basis selection in ACE MLIPs, producing models with improved efficiency, generalization accuracy, and interpretability on benchmark datasets.
Regression shrinkage and selection via the lasso
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.comp-ph 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Scalable Data-Driven Basis Selection for Linear Machine Learning Interatomic Potentials
Active set algorithms enable automated data-driven sparse basis selection in ACE MLIPs, producing models with improved efficiency, generalization accuracy, and interpretability on benchmark datasets.