Incorporation of density scaling constraint in density functional design via contrastive representation learning

Abstract

In a data-driven paradigm, machine learning (ML) is the central component for developing accurate and universal exchange–correlation (XC) functionals in density functional theory (DFT). It is well known that XC functionals must satisfy several exact conditions and physical constraints, such as density scaling, spin scaling, and derivative discontinuity. However, these physical constraints are generally not incorporated implicitly into machine learning through model design or pre-processing on large material datasets. In this work, we demonstrate that contrastive learning is a computationally efficient and flexible method to incorporate a physical constraint, especially when the constraint is defined by an equality, in ML-based density functional design. We propose a schematic approach to incorporate the uniform density scaling property of electron density for exchange energies by adopting contrastive representation learning during the pretraining task. The pretrained hidden representation is transferred to the downstream task to predict the exchange energies calculated by DFT. Based on the computed electron density and exchange energies of around 10 000 molecules in the QM9 database, the augmented molecular density dataset is generated using the density scaling property of exchange energy functionals based on the chosen scaling factors. The electron density encoder transferred from the pretraining task based on contrastive learning predicts exchange energies that satisfy the scaling property, while the model trained without using contrastive learning gives poor predictions for the scaling-transformed electron density systems. Furthermore, the model with pretrained encoder gives satisfactory performance with only small fractions of the whole augmented dataset labeled, comparable to the model trained from scratch using the whole dataset. The results demonstrate that incorporating exact constraints through contrastive learning can enhance the understanding of density-energy mapping using neural network (NN) models with less data labeling, which will be beneficial to generalize the application of NN-based XC functionals in a wide range of scenarios which are not always available experimentally but are theoretically available and justified. This work represents a viable pathway toward the machine learning design of a universal density functional via representation learning.