Search for: All records

New functionality is added to the LAMMPS molecular simulation package, which increases the versatility with which LAMMPS can interface with supporting software and manipulate information associated with bonded force fields. We introduce the “type label” framework that allows atom types and their higher-order interactions (bonds, angles, dihedrals, and impropers) to be represented in terms of the standard atom type strings of a bonded force field. Type labels increase the human readability of input files, enable bonded force fields to be supported by the OpenKIM repository, simplify the creation of reaction templates for the REACTER protocol, and increase compatibility with external visualization tools, such as VMD and OVITO. An introductory primer on the forms and use of bonded force fields is provided to motivate this new functionality and serve as an entry point for LAMMPS and OpenKIM users unfamiliar with bonded force fields. The type label framework has the potential to streamline modeling workflows that use LAMMPS by increasing the portability of software, files, and scripts for preprocessing, running, and postprocessing a molecular simulation. 

For decades, atomistic modeling has played a crucial role in predicting the behavior of materials in numerous fields ranging from nanotechnology to drug discovery. The most accurate methods in this domain are rooted in first-principles quantum mechanical calculations such as density functional theory (DFT). Because these methods have remained computationally prohibitive, practitioners have traditionally focused on defining physically motivated closed-form expressions known as empirical interatomic potentials (EIPs) that approximately model the interactions between atoms in materials. In recent years, neural network (NN)-based potentials trained on quantum mechanical (DFT-labeled) data have emerged as a more accurate alternative to conventional EIPs. However, the generalizability of these models relies heavily on the amount of labeled training data, which is often still insufficient to generate models suitable for general-purpose applications. In this paper, we propose two generic strategies that take advantage of unlabeled training instances to inject domain knowledge from conventional EIPs to NNs in order to increase their generalizability. The first strategy, based on weakly supervised learning, trains an auxiliary classifier on EIPs and selects the best-performing EIP to generate energies to supplement the ground-truth DFT energies in training the NN. The second strategy, based on transfer learning, first pretrains the NN on a large set of easily obtainable EIP energies, and then fine-tunes it on ground-truth DFT energies. Experimental results on three benchmark datasets demonstrate that the first strategy improves baseline NN performance by 5% to 51% while the second improves baseline performance by up to 55%. Combining them further boosts performance.