Adeesh Kolluru
Research Subjects
Adeesh Kolluru has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
- Abhijeet Sadashiv Gangan
Department of Materials Science and Engineering, University of California, Los Angeles, USA
1 shared publication - Abhishek Das
Fundamental AI Research (FAIR), Meta AI, Meta, Menlo Park, USA
1 shared publication Aini Palizhati
Department of Chemical Engineering, Carnegie Mellon University 6 , Pittsburgh, Pennsylvania 15213, USA
1 shared publicationAlexander E. Cohen
Department of Mathematics, Massachusetts Institute of Technology 3 , Cambridge, Massachusetts 02142, USA
1 shared publication- Amalie Trewartha
Toyota Research Institute, Los Altos, California, USA
1 shared publication - Arash Khajeh
Toyota Research Institute, Los Altos, USA
1 shared publication
Latest Publications
- Open Access
- 14 November 2025
- Published by IOP Publishing in AI for Science
- Vol. 1 (2) , 025003
TorchSim: an efficient atomistic simulation engine in PyTorch
We introduce TorchSim, an open-source atomistic simulation engine tailored for the machine learned interatomic potential (MLIP) era. By rewriting core atomistic simulation primitives in PyTorch, TorchSim can achieve orders of magnitude acceleration for popular MLIPs. Unlike existing molecular dynamics (MD) packages, which simulate one system at a time, TorchSim performs batched simulations that efficiently utilize modern GPUs by evolving multiple systems concurrently. TorchSim supports MD integrators, structural relaxation optimizers, both machine-learned and classical interatomic potentials (such as Lennard–Jones, Morse, soft-sphere), batching with automatic memory management, differentiable simulation, and integration with popular materials informatics tools.
- 14 November 2025
- Published by IOP Publishing in AI for Science
- Vol. 1 (2) , 025003
Open AccessTorchSim: an efficient atomistic simulation engine in PyTorch
We introduce TorchSim, an open-source atomistic simulation engine tailored for the machine learned interatomic potential (MLIP) era. By rewriting core atomistic simulation primitives in PyTorch, TorchSim can achieve orders of magnitude acceleration for popular MLIPs. Unlike existing molecular dynamics (MD) packages, which simulate one system at a time, TorchSim performs batched simulations that efficiently utilize modern GPUs by evolving multiple systems concurrently. TorchSim supports MD integrators, structural relaxation optimizers, both machine-learned and classical interatomic potentials (such as Lennard–Jones, Morse, soft-sphere), batching with automatic memory management, differentiable simulation, and integration with popular materials informatics tools.
- Open Access
- 1 June 2023
- Published by AIP Publishing in APL Machine Learning
- Vol. 1 (2)
Materials cartography: A forward-looking perspective on materials representation and devising better maps
Machine learning (ML) is gaining popularity as a tool for materials scientists to accelerate computation, automate data analysis, and predict materials properties. The representation of input material features is critical to the accuracy, interpretability, and generalizability of data-driven models for scientific research. In this Perspective, we discuss a few central challenges faced by ML practitioners in developing meaningful representations, including handling the complexity of real-world industry-relevant materials, combining theory and experimental data sources, and describing scientific phenomena across timescales and length scales. We present several promising directions for future research: devising representations of varied experimental conditions and observations, the need to find ways to integrate machine learning into laboratory practices, and making multi-scale informatics toolkits to bridge the gaps between atoms, materials, and devices.
- 1 June 2023
- Published by AIP Publishing in APL Machine Learning
- Vol. 1 (2)
Open AccessMaterials cartography: A forward-looking perspective on materials representation and devising better maps
Machine learning (ML) is gaining popularity as a tool for materials scientists to accelerate computation, automate data analysis, and predict materials properties. The representation of input material features is critical to the accuracy, interpretability, and generalizability of data-driven models for scientific research. In this Perspective, we discuss a few central challenges faced by ML practitioners in developing meaningful representations, including handling the complexity of real-world industry-relevant materials, combining theory and experimental data sources, and describing scientific phenomena across timescales and length scales. We present several promising directions for future research: devising representations of varied experimental conditions and observations, the need to find ways to integrate machine learning into laboratory practices, and making multi-scale informatics toolkits to bridge the gaps between atoms, materials, and devices.
- 12 May 2022
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 156 (18) , 184702
Transfer learning using attentions across atomic systems with graph neural networks (TAAG)
Recent advances in Graph Neural Networks (GNNs) have transformed the space of molecular and catalyst discovery. Even though the underlying physics across these domains remain the same, most prior work has focused on building domain-specific models either in small molecules or in materials. However, building large datasets across domains is computationally expensive, therefore the use of transfer learning (TL) to generalize to different domains is a promising but under-explored approach to this problem. To evaluate this hypothesis, we use a model that is pre-trained on the Open Catalyst Dataset (0C20), and we study the model's behavior when fine-tuned for a set of different datasets and tasks. This includes MD17, CO adsorbate dataset, and across tasks in OC20. Through extensive TL experiments, we demonstrate that initial layers of GNNs learn more basic representation that holds across domains, whereas final layers learn more task-specific features. Moreover, these well-known strategies give significant improvement over the non-pretrained models for in-domain tasks with improvements of 53% and 17% for CO dataset and across OCP task respectively. TL approaches also gives up to 4x speedup in model training depending on the target data and task. However, this does not perform well for MD17 due and gives worse performance than non-pretrained model for few molecules. Based on these observations, we propose TAAG, an attention-based approach that adapts to transfer important features from the interaction layers of GNNs. This method outperforms the best TL approach for MD17 and gives a mean improvement of 6% over the non-pretrained model.
- 12 May 2022
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 156 (18) , 184702
Transfer learning using attentions across atomic systems with graph neural networks (TAAG)
Recent advances in Graph Neural Networks (GNNs) have transformed the space of molecular and catalyst discovery. Even though the underlying physics across these domains remain the same, most prior work has focused on building domain-specific models either in small molecules or in materials. However, building large datasets across domains is computationally expensive, therefore the use of transfer learning (TL) to generalize to different domains is a promising but under-explored approach to this problem. To evaluate this hypothesis, we use a model that is pre-trained on the Open Catalyst Dataset (0C20), and we study the model's behavior when fine-tuned for a set of different datasets and tasks. This includes MD17, CO adsorbate dataset, and across tasks in OC20. Through extensive TL experiments, we demonstrate that initial layers of GNNs learn more basic representation that holds across domains, whereas final layers learn more task-specific features. Moreover, these well-known strategies give significant improvement over the non-pretrained models for in-domain tasks with improvements of 53% and 17% for CO dataset and across OCP task respectively. TL approaches also gives up to 4x speedup in model training depending on the target data and task. However, this does not perform well for MD17 due and gives worse performance than non-pretrained model for few molecules. Based on these observations, we propose TAAG, an attention-based approach that adapts to transfer important features from the interaction layers of GNNs. This method outperforms the best TL approach for MD17 and gives a mean improvement of 6% over the non-pretrained model.