Justin Steven Smith
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Justin Steven Smith has a total of 126 co-authors. Here are the top co-authors by number of shared publications:
- Benjamin Nebgen
Los Alamos National Laboratory
16 shared publications - Sergei Tretiak
Theoretical Division, Los Alamos National Laboratory
12 shared publications - Nicholas Lubbers
Los Alamos National Laboratory
11 shared publications - Kipton Barros
Los Alamos National Laboratory
10 shared publications - Olexandr Isayev
Carnegie Mellon University
10 shared publications - Richard Alma Messerly
National Center for Computational Sciences Division, Oak Ridge National Laboratory, Oak Ridge, 37830, TN, USA
7 shared publications
Latest Publications
- Open Access
- 29 September 2025
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 6 (3) , 035066
Multi-fidelity learning for interatomic potentials: low-level forces and high-level energies are all you need*
The promise of machine learning interatomic potentials (MLIPs) has led to an abundance of public quantum mechanical (QM) training datasets. The quality of an MLIP is directly limited by the accuracy of the energies and atomic forces in the training dataset. Unfortunately, most of these datasets are computed with relatively low-accuracy QM methods, e.g. density functional theory with a moderate basis set. Due to the increased computational cost of more accurate QM methods, e.g. coupled-cluster theory with a complete basis set (CBS) extrapolation, most high-accuracy datasets are much smaller and often do not contain atomic forces. The lack of high-accuracy atomic forces is quite troubling, as training with force data greatly improves the stability and quality of the MLIP compared to training to energy alone. Because most datasets are computed with a unique level of theory, traditional single-fidelity (SF) learning is not capable of leveraging the vast amounts of published QM data. In this study, we apply multi-fidelity learning (MFL) to train an MLIP to multiple QM datasets of different levels of accuracy, i.e. levels of fidelity. Specifically, we perform three test cases to demonstrate that MFL with both low-level forces and high-level energies yields an extremely accurate MLIP—far more accurate than a SF MLIP trained solely to high-level energies and almost as accurate as a SF MLIP trained directly to high-level energies and forces. Therefore, MFL greatly alleviates the need for generating large and expensive datasets containing high-accuracy atomic forces and allows for more effective training to existing high-accuracy energy-only datasets. Indeed, low-accuracy atomic forces and high-accuracy energies are all that are needed to achieve a high-accuracy MLIP with MFL.
- 29 September 2025
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 6 (3) , 035066
Open AccessMulti-fidelity learning for interatomic potentials: low-level forces and high-level energies are all you need*
The promise of machine learning interatomic potentials (MLIPs) has led to an abundance of public quantum mechanical (QM) training datasets. The quality of an MLIP is directly limited by the accuracy of the energies and atomic forces in the training dataset. Unfortunately, most of these datasets are computed with relatively low-accuracy QM methods, e.g. density functional theory with a moderate basis set. Due to the increased computational cost of more accurate QM methods, e.g. coupled-cluster theory with a complete basis set (CBS) extrapolation, most high-accuracy datasets are much smaller and often do not contain atomic forces. The lack of high-accuracy atomic forces is quite troubling, as training with force data greatly improves the stability and quality of the MLIP compared to training to energy alone. Because most datasets are computed with a unique level of theory, traditional single-fidelity (SF) learning is not capable of leveraging the vast amounts of published QM data. In this study, we apply multi-fidelity learning (MFL) to train an MLIP to multiple QM datasets of different levels of accuracy, i.e. levels of fidelity. Specifically, we perform three test cases to demonstrate that MFL with both low-level forces and high-level energies yields an extremely accurate MLIP—far more accurate than a SF MLIP trained solely to high-level energies and almost as accurate as a SF MLIP trained directly to high-level energies and forces. Therefore, MFL greatly alleviates the need for generating large and expensive datasets containing high-accuracy atomic forces and allows for more effective training to existing high-accuracy energy-only datasets. Indeed, low-accuracy atomic forces and high-accuracy energies are all that are needed to achieve a high-accuracy MLIP with MFL.
- Open Access
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Optimizing Data Distribution and Kernel Performance for Efficient Training of Chemistry Foundation Models: A Case Study with MACE
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessOptimizing Data Distribution and Kernel Performance for Efficient Training of Chemistry Foundation Models: A Case Study with MACE
- Open Access
Optimizing Data Distribution and Kernel Performance for Efficient Training of Chemistry Foundation Models: A Case Study with MACE
Chemistry Foundation Models (CFMs) that leverage Graph Neural Networks (GNNs) operating on 3D molecular graph structures are becoming indispensable tools for computational chemists and materials scientists. These models facilitate the understanding of matter and the discovery of new molecules and materials. In contrast to GNNs operating on a large homogeneous graphs, GNNs used by CFMs process a large number of geometric graphs of varying sizes, requiring different optimization strategies than those developed for large homogeneous GNNs. This paper presents optimizations for two critical phases of CFM training: data distribution and model training, targeting MACE - a state-of-the-art CFM. We address the challenge of load balancing in data distribution by formulating it as a multi-objective bin packing problem. We propose an iterative algorithm that provides a highly effective, fast, and practical solution, ensuring efficient data distribution. For the training phase, we identify symmetric tensor contraction as the key computational kernel in MACE and optimize this kernel to improve the overall performance. Our combined approach of balanced data distribution and kernel optimization significantly enhances the training process of MACE. Experimental results demonstrate a substantial speedup, reducing per-epoch execution time for training from 12 to 2 minutes on 740 GPUs with a 2.6M sample dataset.
Open AccessOptimizing Data Distribution and Kernel Performance for Efficient Training of Chemistry Foundation Models: A Case Study with MACE
Chemistry Foundation Models (CFMs) that leverage Graph Neural Networks (GNNs) operating on 3D molecular graph structures are becoming indispensable tools for computational chemists and materials scientists. These models facilitate the understanding of matter and the discovery of new molecules and materials. In contrast to GNNs operating on a large homogeneous graphs, GNNs used by CFMs process a large number of geometric graphs of varying sizes, requiring different optimization strategies than those developed for large homogeneous GNNs. This paper presents optimizations for two critical phases of CFM training: data distribution and model training, targeting MACE - a state-of-the-art CFM. We address the challenge of load balancing in data distribution by formulating it as a multi-objective bin packing problem. We propose an iterative algorithm that provides a highly effective, fast, and practical solution, ensuring efficient data distribution. For the training phase, we identify symmetric tensor contraction as the key computational kernel in MACE and optimize this kernel to improve the overall performance. Our combined approach of balanced data distribution and kernel optimization significantly enhances the training process of MACE. Experimental results demonstrate a substantial speedup, reducing per-epoch execution time for training from 12 to 2 minutes on 740 GPUs with a 2.6M sample dataset.
- Top Cited PapersOpen Access
- 7 March 2024
- Published by Springer Nature in Nature Chemistry
- Vol. 16 (5) , 727-734
Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential
Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.
- 7 March 2024
- Published by Springer Nature in Nature Chemistry
- Vol. 16 (5) , 727-734
Top Cited PapersOpen AccessExploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential
Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.
- Open Access
- 28 June 2023
- Published by Springer Nature in Research Square
Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential
Reactive chemistry atomistic simulation has a broad range of applications from drug design to energy to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive quantum chemistry simulations. In practice, developing reactive MLIPs requires prior knowledge of reaction networks to generate fitting data and refitting to extensive datasets for each new application. For this reason, many fields of chemistry would greatly benefit from a general reactive MLIP, i.e., an MLIP that is applicable to a broad range of reactive chemistry such that it can be applied to new systems without the need for retraining. In this work, we develop a general reactive MLIP through unbiased active learning with an atomic configuration sampler inspired by nanoreactor molecular dynamics. The resulting potential (ANI-1xnr) is then applied to study five distinct condensed-phase reactive chemistry systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early-earth small molecules. In all studies, ANI-1xnr closely matches experiment and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N, and O elements that does not need to be refit for each application, enabling high-throughput in silico reactive chemistry experimentation.
- 28 June 2023
- Published by Springer Nature in Research Square
Open AccessExploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential
Reactive chemistry atomistic simulation has a broad range of applications from drug design to energy to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive quantum chemistry simulations. In practice, developing reactive MLIPs requires prior knowledge of reaction networks to generate fitting data and refitting to extensive datasets for each new application. For this reason, many fields of chemistry would greatly benefit from a general reactive MLIP, i.e., an MLIP that is applicable to a broad range of reactive chemistry such that it can be applied to new systems without the need for retraining. In this work, we develop a general reactive MLIP through unbiased active learning with an atomic configuration sampler inspired by nanoreactor molecular dynamics. The resulting potential (ANI-1xnr) is then applied to study five distinct condensed-phase reactive chemistry systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early-earth small molecules. In all studies, ANI-1xnr closely matches experiment and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N, and O elements that does not need to be refit for each application, enabling high-throughput in silico reactive chemistry experimentation.
- Open Access
- 9 May 2023
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 158 (18)
Lightweight and effective tensor sensitivity for atomistic neural networks
Atomistic machine learning focuses on the creation of models that obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here, we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 kcalmol for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature.
- 9 May 2023
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 158 (18)
Open AccessLightweight and effective tensor sensitivity for atomistic neural networks
Atomistic machine learning focuses on the creation of models that obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here, we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 kcalmol for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature.
- Open Access
- 16 April 2023
- Published by American Chemical Society (ACS) in ChemRxiv
Exploring the frontiers of chemistry with a generalreactive machine learning potential
Reactive chemistry atomistic simulation has a broad range of applications from drug design to energy to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive quantum chemistry simulations. In practice, developing reactive MLIPs requires prior knowledge of reaction networks to generate fitting data and refitting to extensive datasets for each new application. In this work, we develop a general reactive MLIP through unbiased active learning with an atomic configuration sampler inspired by nanoreactor molecular dynamics. The resulting potential (ANI-1xnr) is then applied to study five distinct condensed phase reactive chemistry systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early-earth small molecules. In all studies, ANI-1xnr closely matches experiment and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP that does not need to be refit for each application, enabling high-throughput in silico reactive chemistry experimentation.
- 16 April 2023
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessExploring the frontiers of chemistry with a generalreactive machine learning potential
Reactive chemistry atomistic simulation has a broad range of applications from drug design to energy to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive quantum chemistry simulations. In practice, developing reactive MLIPs requires prior knowledge of reaction networks to generate fitting data and refitting to extensive datasets for each new application. In this work, we develop a general reactive MLIP through unbiased active learning with an atomic configuration sampler inspired by nanoreactor molecular dynamics. The resulting potential (ANI-1xnr) is then applied to study five distinct condensed phase reactive chemistry systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early-earth small molecules. In all studies, ANI-1xnr closely matches experiment and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP that does not need to be refit for each application, enabling high-throughput in silico reactive chemistry experimentation.
- Top Cited PapersOpen Access
- 6 March 2023
- Published by Springer Nature in Nature Computational Science
- Vol. 3 (3) , 230-239
Uncertainty-driven dynamics for active learning of interatomic potentials
Machine learning (ML) models, if trained to data sets of high-fidelity quantum simulations, produce accurate and efficient interatomic potentials. Active learning (AL) is a powerful tool to iteratively generate diverse data sets. In this approach, the ML model provides an uncertainty estimate along with its prediction for each new atomic configuration. If the uncertainty estimate passes a certain threshold, then the configuration is included in the data set. Here we develop a strategy to more rapidly discover configurations that meaningfully augment the training data set. The approach, uncertainty-driven dynamics for active learning (UDD-AL), modifies the potential energy surface used in molecular dynamics simulations to favor regions of configuration space for which there is large model uncertainty. The performance of UDD-AL is demonstrated for two AL tasks: sampling the conformational space of glycine and sampling the promotion of proton transfer in acetylacetone. The method is shown to efficiently explore the chemically relevant configuration space, which may be inaccessible using regular dynamical sampling at target temperature conditions.
- 6 March 2023
- Published by Springer Nature in Nature Computational Science
- Vol. 3 (3) , 230-239
Top Cited PapersOpen AccessUncertainty-driven dynamics for active learning of interatomic potentials
Machine learning (ML) models, if trained to data sets of high-fidelity quantum simulations, produce accurate and efficient interatomic potentials. Active learning (AL) is a powerful tool to iteratively generate diverse data sets. In this approach, the ML model provides an uncertainty estimate along with its prediction for each new atomic configuration. If the uncertainty estimate passes a certain threshold, then the configuration is included in the data set. Here we develop a strategy to more rapidly discover configurations that meaningfully augment the training data set. The approach, uncertainty-driven dynamics for active learning (UDD-AL), modifies the potential energy surface used in molecular dynamics simulations to favor regions of configuration space for which there is large model uncertainty. The performance of UDD-AL is demonstrated for two AL tasks: sampling the conformational space of glycine and sampling the promotion of proton transfer in acetylacetone. The method is shown to efficiently explore the chemically relevant configuration space, which may be inaccessible using regular dynamical sampling at target temperature conditions.
- Open Access
- 16 November 2022
- Published by Springer Nature in Nature Reviews Chemistry
- Vol. 6 (12) , 916
Publisher Correction: Extending machine learning beyond interatomic potentials for predicting molecular properties
- 16 November 2022
- Published by Springer Nature in Nature Reviews Chemistry
- Vol. 6 (12) , 916
Open AccessPublisher Correction: Extending machine learning beyond interatomic potentials for predicting molecular properties
- Top Cited Papers
- 25 August 2022
- Published by Springer Nature in Nature Reviews Chemistry
- Vol. 6 (9) , 653-672
Extending machine learning beyond interatomic potentials for predicting molecular properties
Machine learning (ML) is becoming a method of choice for modelling complex chemical processes and materials. ML provides a surrogate model trained on a reference dataset that can be used to establish a relationship between a molecular structure and its chemical properties. This Review highlights developments in the use of ML to evaluate chemical properties such as partial atomic charges, dipole moments, spin and electron densities, and chemical bonding, as well as to obtain a reduced quantum-mechanical description. We overview several modern neural network architectures, their predictive capabilities, generality and transferability, and illustrate their applicability to various chemical properties. We emphasize that learned molecular representations resemble quantum-mechanical analogues, demonstrating the ability of the models to capture the underlying physics. We also discuss how ML models can describe non-local quantum effects. Finally, we conclude by compiling a list of available ML toolboxes, summarizing the unresolved challenges and presenting an outlook for future development. The observed trends demonstrate that this field is evolving towards physics-based models augmented by ML, which is accompanied by the development of new methods and the rapid growth of user-friendly ML frameworks for chemistry.
- 25 August 2022
- Published by Springer Nature in Nature Reviews Chemistry
- Vol. 6 (9) , 653-672
Top Cited PapersExtending machine learning beyond interatomic potentials for predicting molecular properties
Machine learning (ML) is becoming a method of choice for modelling complex chemical processes and materials. ML provides a surrogate model trained on a reference dataset that can be used to establish a relationship between a molecular structure and its chemical properties. This Review highlights developments in the use of ML to evaluate chemical properties such as partial atomic charges, dipole moments, spin and electron densities, and chemical bonding, as well as to obtain a reduced quantum-mechanical description. We overview several modern neural network architectures, their predictive capabilities, generality and transferability, and illustrate their applicability to various chemical properties. We emphasize that learned molecular representations resemble quantum-mechanical analogues, demonstrating the ability of the models to capture the underlying physics. We also discuss how ML models can describe non-local quantum effects. Finally, we conclude by compiling a list of available ML toolboxes, summarizing the unresolved challenges and presenting an outlook for future development. The observed trends demonstrate that this field is evolving towards physics-based models augmented by ML, which is accompanied by the development of new methods and the rapid growth of user-friendly ML frameworks for chemistry.