mlsanityData to 14 Oct 2026

Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks

arXiv 2111.15176 ↗First posted 30 Nov 2021physics.comp-ph cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Molecular dynamics (MD) simulation predicts the trajectory of atoms by solving Newton's equation of motion with a numeric integrator. Due to physical constraints, the time step of the integrator need to be small to maintain sufficient precision. This limits the efficiency of simulation. To this end, we introduce a graph neural network (GNN) based model, MDNet, to predict the evolution of coordinates and momentum with large time steps. In addition, MDNet can easily scale to a larger system, due to its linear complexity with respect to the system size. We demonstrate the performance of MDNet on a 4000-atom system with large time steps, and show that MDNet can predict good equilibrium and transport properties, well aligned with standard MD simulations.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 11 of its 31 references that are in the map. First posted on arXiv 30 Nov 2021.

Builds on

11 of 31 references.

References · 11

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2021E(n) Equivariant Graph Neural Networkscited by 965
  3. 2019Hamiltonian Neural Networkscited by 712
  4. 2017Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanicscited by 214
  5. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 212
  6. 2019Symplectic ODE-Net: Learning Hamiltonian Dynamics with Controlcited by 184
  7. 2019Symplectic Recurrent Neural Networkscited by 150
  8. 2019Hamiltonian Generative Networkscited by 137
  9. 2021Data-driven Prediction of General Hamiltonian Dynamics via Learning Exactly-Symplectic Mapscited by 41
  10. 2019Structure-preserving Method for Reconstructing Unknown Hamiltonian Systems from Trajectory Datacited by 18
  11. 2018Machine learning enables long time scale molecular photodynamics simulationscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 5; 2026: 2.

9 papers.

  1. 2025FlashMD: long-stride, universal prediction of molecular dynamicscited by 12
  2. 2026Learning Hamiltonian Flow Maps: Mean Flow Consistency for Large-Timestep Molecular Dynamicscited by 7
  3. 2023Evaluating the Transferability of Machine-Learned Force Fields for Material Property Modelingcited by 2
  4. 2025Artificial Intelligence for Direct Prediction of Molecular Dynamics Across Chemical Spacecited by 2
  5. 2025Learning the action for long-time-step simulations of molecular dynamicscited by 2
  6. 2025Accelerating Long-Term Molecular Dynamics with Physics-Informed Time-Series Forecastingcited by 2
  7. 2025ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamicscited by 1
  8. 2024Generalizability of Graph Neural Network Force Fields for Predicting Solid-State Propertiescited by 0
  9. 2026Teaching Molecular Dynamics to a Non-Autoregressive Ionic Transport Predictorcited by 0