Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids
This work addresses the computational bottleneck in electronic structure calculations for molecules and solids, offering a transferable model to accelerate large-scale simulations, though it builds on existing equivariant frameworks.
The authors tackled the inefficiency of density functional theory (DFT) calculations by using a message-passing machine learning approach to directly map structures to electronic Hamiltonian matrices, achieving state-of-the-art precision in benchmarks and accurately predicting properties for various systems.
Using the message-passing mechanism in machine learning (ML) instead of self-consistent iterations to directly build the mapping from structures to electronic Hamiltonian matrices will greatly improve the efficiency of density functional theory (DFT) calculations. In this work, we proposed a general analytic Hamiltonian representation in an E(3) equivariant framework, which can fit the ab initio Hamiltonian of molecules and solids by a complete data-driven method and are equivariant under rotation, space inversion, and time reversal operations. Our model reached state-of-the-art precision in the benchmark test and accurately predicted the electronic Hamiltonian matrices and related properties of various periodic and aperiodic systems, showing high transferability and generalization ability. This framework provides a general transferable model that can be used to accelerate the electronic structure calculations on different large systems with the same network weights trained on small structures.