Scaling the leading accuracy of deep equivariant models to biomolecular simulations of realistic size

• URL: http://arxiv.org/abs/2304.10061v1
• Date: Thu, 20 Apr 2023 03:02:25 GMT
• Title: Scaling the leading accuracy of deep equivariant models to biomolecular simulations of realistic size
• Authors: Albert Musaelian, Anders Johansson, Simon Batzner, Boris Kozinsky
• Abstract summary: This work brings the leading accuracy, sample efficiency, and robustness of deep equivariant neural networks to the extreme computational scale. It is achieved through a combination of innovative model architecture, massive parallelization, and models and implementations optimized for efficient GPU utilization.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This work brings the leading accuracy, sample efficiency, and robustness of deep equivariant neural networks to the extreme computational scale. This is achieved through a combination of innovative model architecture, massive parallelization, and models and implementations optimized for efficient GPU utilization. The resulting Allegro architecture bridges the accuracy-speed tradeoff of atomistic simulations and enables description of dynamics in structures of unprecedented complexity at quantum fidelity. To illustrate the scalability of Allegro, we perform nanoseconds-long stable simulations of protein dynamics and scale up to a 44-million atom structure of a complete, all-atom, explicitly solvated HIV capsid on the Perlmutter supercomputer. We demonstrate excellent strong scaling up to 100 million atoms and 70% weak scaling to 5120 A100 GPUs.

Related papers

• A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures [73.65190161312555]
  ARCANA is a spiking neural network simulator designed to account for the properties of mixed-signal neuromorphic circuits. We show how the results obtained provide a reliable estimate of the behavior of the spiking neural network trained in software.
  arXiv  Detail & Related papers  (2024-09-23T11:16:46Z)
• Mechanistic Design and Scaling of Hybrid Architectures [114.3129802943915]
  We identify and test new hybrid architectures constructed from a variety of computational primitives. We experimentally validate the resulting architectures via an extensive compute-optimal and a new state-optimal scaling law analysis. We find MAD synthetics to correlate with compute-optimal perplexity, enabling accurate evaluation of new architectures.
  arXiv  Detail & Related papers  (2024-03-26T16:33:12Z)
• A Multi-Grained Symmetric Differential Equation Model for Learning Protein-Ligand Binding Dynamics [74.93549765488103]
  In drug discovery, molecular dynamics simulation provides a powerful tool for predicting binding affinities, estimating transport properties, and exploring pocket sites. We propose NeuralMD, the first machine learning surrogate that can facilitate numerical MD and provide accurate simulations in protein-ligand binding. We show the efficiency and effectiveness of NeuralMD, with a 2000$times$ speedup over standard numerical MD simulation and outperforming all other ML approaches by up to 80% under the stability metric.
  arXiv  Detail & Related papers  (2024-01-26T09:35:17Z)
• Rethinking materials simulations: Blending direct numerical simulations with neural operators [1.6874375111244329]
  We develop a new method that blends numerical solvers with neural operators to accelerate such simulations. We demonstrate the effectiveness of this framework on simulations of microstructure evolution during physical vapor deposition.
  arXiv  Detail & Related papers  (2023-12-08T23:44:54Z)
• Navigating protein landscapes with a machine-learned transferable coarse-grained model [29.252004942896875]
  coarse-grained (CG) model with similar prediction performance has been a long-standing challenge. We develop a bottom-up CG force field with chemical transferability, which can be used for extrapolative molecular dynamics on new sequences. We demonstrate that the model successfully predicts folded structures, intermediates, metastable folded and unfolded basins, and the fluctuations of intrinsically disordered proteins.
  arXiv  Detail & Related papers  (2023-10-27T17:10:23Z)
• Efficient Quantum Circuit Simulation by Tensor Network Methods on Modern GPUs [11.87665112550076]
  In quantum hardware, primary simulation methods are based on state vectors and tensor networks. As the number of qubits and quantum gates grows larger, traditional state-vector based quantum circuit simulation methods prove inadequate due to the overwhelming size of the Hilbert space and extensive entanglement. In this study, we propose general optimization strategies from two aspects: computational efficiency and accuracy.
  arXiv  Detail & Related papers  (2023-10-06T02:24:05Z)
• Allegro-Legato: Scalable, Fast, and Robust Neural-Network Quantum Molecular Dynamics via Sharpness-Aware Minimization [1.8431330466822737]
  Neural-network quantum molecular dynamics (NNQMD) simulations based on machine learning are revolutionizing atomistic simulations of materials by providing quantum-mechanical accuracy but orders-of-magnitude faster. State-of-the-art (SOTA) NNQMD model founded on group theory featuring rotational equivariance and local descriptors has provided much higher accuracy and speed than those models, thus named Allegro (meaning fast) On massively parallel supercomputers, however, it suffers a fidelity-scaling problem, where growing number of unphysical predictions of interatomic forces prohibits simulations involving larger numbers of atoms for longer times. We
  arXiv  Detail & Related papers  (2023-03-14T18:36:44Z)
• ViSNet: an equivariant geometry-enhanced graph neural network with vector-scalar interactive message passing for molecules [69.05950120497221]
  We propose an equivariant geometry-enhanced graph neural network called ViSNet, which elegantly extracts geometric features and efficiently models molecular structures. Our proposed ViSNet outperforms state-of-the-art approaches on multiple MD benchmarks, including MD17, revised MD17 and MD22, and achieves excellent chemical property prediction on QM9 and Molecule3D datasets.
  arXiv  Detail & Related papers  (2022-10-29T07:12:46Z)
• Multi-fidelity Hierarchical Neural Processes [79.0284780825048]
  Multi-fidelity surrogate modeling reduces the computational cost by fusing different simulation outputs. We propose Multi-fidelity Hierarchical Neural Processes (MF-HNP), a unified neural latent variable model for multi-fidelity surrogate modeling. We evaluate MF-HNP on epidemiology and climate modeling tasks, achieving competitive performance in terms of accuracy and uncertainty estimation.
  arXiv  Detail & Related papers  (2022-06-10T04:54:13Z)
• Learning Local Equivariant Representations for Large-Scale Atomistic Dynamics [0.6861083714313458]
  Allegro is a strictly local equivariant deep learning interatomic potential. It simultaneously exhibits excellent accuracy and scalability of parallel computation. A single tensor product layer is shown to outperform existing deep message passing neural networks and transformers.
  arXiv  Detail & Related papers  (2022-04-11T16:48:41Z)
• Using Machine Learning at Scale in HPC Simulations with SmartSim: An Application to Ocean Climate Modeling [52.77024349608834]
  We demonstrate the first climate-scale, numerical ocean simulations improved through distributed, online inference of Deep Neural Networks (DNN) using SmartSim. SmartSim is a library dedicated to enabling online analysis and Machine Learning (ML) for traditional HPC simulations.
  arXiv  Detail & Related papers  (2021-04-13T19:27:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.