Exact constraints and appropriate norms in machine learned exchange-correlation functionals

Abstract

Machine learning techniques have received growing attention as an alternative strategy for developing general-purpose density functional approximations, augmenting the historically successful approach of human designed functionals derived to obey mathematical constraints known for the exact exchange-correlation functional. More recently efforts have been made to reconcile the two techniques, integrating machine learning and exact-constraint satisfaction. We continue this integrated approach, designing a deep neural network that exploits the exact constraint and appropriate norm philosophy to deorbitalize the strongly constrained and appropriately normed SCAN functional. The deep neural network is trained to replicate the SCAN functional from only electron density and local derivative information, avoiding use of the orbital dependent kinetic energy density. The performance and transferability of the machine learned functional are demonstrated for molecular and periodic systems.

References in corpus (25)

• Adam: A Method for Stochastic Optimization
• Scikit-learn: Machine Learning in Python
• TensorFlow: A system for large-scale machine learning
• Climbing the Density Functional Ladder: Non-Empirical Meta-Generalized Gradient Approximation Designed for Molecules and Solids
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• By-passing the Kohn-Sham equations with machine learning
• Finding Density Functionals with Machine Learning
• Ab initio theory and modeling of water
• Regularized SCAN functional
• More-Realistic Band Gaps from Meta-Generalized Gradient Approximations: Only in a Generalized Kohn-Sham Scheme
• Differentiable Programming Tensor Networks
• Perspective on integrating machine learning into computational chemistry and materials science
• Density Functionals that Recognize Covalent, Metallic, and Weak Bonds
• Evaluating transition-metal oxides within DFT-SCAN and SCAN+U frameworks for solar thermochemical applications
• Orbital-free Bond Breaking via Machine Learning
• Antiferromagnetic Ground State of La${}_2$CuO${}_4$: A Parameter-free Ab Initio Description
• Pure density functional for strong correlations and the thermodynamic limit from machine learning
• Gedanken Densities and Exact Constraints in Density Functional Theory
• Deorbitalized meta-GGA Exchange-Correlation Functionals in Solids
• Accurate and numerically efficient r${}^2$SCAN meta-generalized gradient approximation
• Deorbitalization strategies for meta-GGA exchange-correlation functionals
• Neural-network Kohn-Sham exchange-correlation potential and its out-of-training transferability
• Enhancing the efficiency of density functionals with a novel iso-orbital indicator
• Metal Surface Energy: Persistent Cancellation of Short-Range Correlation Effects beyond the Random-Phase Approximation
• Nonempirical Density Functionals Investigated for Jellium: Spin-Polarized Surfaces, Spherical Clusters, and Bulk Linear Response

Cited by in corpus (4)

• DeePKS+ABACUS as a Bridge between Expensive Quantum Mechanical Models and Machine Learning Potentials
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• The Convexity Condition of Density-Functional Theory
• Exchange-Correlation Functionals in 2D Materials: Applications, Challenges, and Limitations