Faster Molecular Dynamics with Neural Network Potentials via Distilled Multiple Time-Stepping and Non-Conservative Forces

Abstract

Following our previous work (J. Phys. Chem. Lett., 2026, 17, 5, 1288-1295), we propose the DMTS-NC approach, a distilled multi-time-step (DMTS) strategy using non-conservative (NC) forces to further accelerate atomistic molecular dynamics simulations using foundation neural network models such as FeNNix-Bio1. There, a dual-level reversible reference system propagator algorithm (RESPA) formalism couples a target accurate conservative potential to a simplified distilled representation optimized for the production of non-conservative forces. Despite being non-conservative, the distilled architecture is designed to enforce key physical priors, such as equivariance under rotation and cancellation of atomic force components. These choices facilitate the distillation process and therefore improve drastically the robustness of simulation, significantly limiting abnormal discrepancies between the two models, thus achieving excellent agreement with the forces data. Overall, the DMTS-NC scheme is found to be more stable and efficient than its conservative counterpart with additional speedups reaching 15-30% over DMTS. Requiring no fine-tuning steps, it is easier to implement and can be pushed to the limit of the systems physical resonances to maintain accuracy while providing maximum efficiency. We obtain additional speedup by combining hydrogen mass repartitioning (HMR), High Hydrogen Friction (HHF) to further extended the largest timestep up to 10fs of our schemes while conserving stability and accuracy. As for DMTS, DMTS-NC is applicable to any neural network potential and can be applied to approaches that are computationally heavier than FeNNix-Bio1. We show a proof of principle applying the approach to the distillation of MACE-OFF23 with consequent speedups ranging from 3.66 to 5.64 compared to single timestep.

References in corpus (24)

• Knowledge Distillation: A Survey
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• DeePMD-kit v2: A software package for Deep Potential models
• Isothermal-isobaric molecular dynamics using stochastic velocity rescaling
• SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects
• Robust and efficient configurational molecular sampling via Langevin Dynamics
• Incorporating long-range physics in atomic-scale machine learning
• Rational Construction of Stochastic Numerical Methods for Molecular Sampling
• Efficient multiple time scale molecular dynamics: using colored noise thermostats to stabilize resonances
• Tinker-HP : Accelerating Molecular Dynamics Simulations of Large Complex Systems with Advanced Point Dipole Polarizable Force Fields using GPUs and Multi-GPUs systems
• SchNetPack 2.0: A neural network toolbox for atomistic machine learning
• Stochastic resonance-free multiple time-step algorithm for molecular dynamics with very large time steps
• MLatom 3: Platform for machine learning-enhanced computational chemistry simulations and workflows
• Scalable Hybrid Deep Neural Networks/Polarizable Potentials Biomolecular Simulations including long-range effects
• Force-Field-Enhanced Neural Network Interactions: from Local Equivariant Embedding to Atom-in-Molecule properties and long-range effects
• Lambda-ABF: Simplified, Portable, Accurate and Cost-effective Alchemical Free Energy Computations
• Repulsive interatomic potentials calculated at three levels of theory
• FeNNol: an Efficient and Flexible Library for Building Force-field-enhanced Neural Network Potentials
• Fast-Forward Langevin Dynamics with Momentum Flips
• Optimal friction matrix for underdamped Langevin sampling
• Velocity Jumps for Molecular Dynamics
• Accelerating Molecular Dynamics Simulations with Foundation Neural Network Models using Multiple Time-Step and Distillation