Enhanced Sampling for Efficient Learning of Coarse-Grained Machine Learning Potentials
Not on the map. Why?
Abstract (arXiv)
Coarse-graining (CG) enables molecular dynamics (MD) simulations of larger systems and longer timescales that are otherwise infeasible with atomistic models. Machine learning potentials (MLPs), with their capacity to capture many-body interactions, can provide accurate approximations of the potential of mean force (PMF) in CG models. Current CG MLPs are typically trained in a bottom-up manner via force matching, which in practice relies on configurations sampled from the unbiased equilibrium Boltzmann distribution to ensure thermodynamic consistency. This convention poses two key limitations: first, sufficiently long atomistic trajectories are needed to reach convergence; and second, even once equilibrated, transition regions remain poorly sampled. To address these issues, we employ enhanced sampling to bias along CG degrees of freedom for data generation, and then recompute the forces with respect to the unbiased potential. This strategy simultaneously shortens the simulation time required to produce equilibrated data and enriches sampling in transition regions, while preserving the correct PMF. We demonstrate its effectiveness on the M"uller-Brown potential and capped alanine, achieving notable improvements. Our findings support the use of enhanced sampling for force matching as a promising direction to improve the accuracy and reliability of CG MLPs.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 37 of its 110 references that are in the map. First posted on arXiv 13 Oct 2025.
Builds on
37 of 110 references.
References · 37
- 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 474
- 2022Flow Annealed Importance Sampling Bootstrapcited by 127
- 2024AlphaFold Meets Flow Matching for Generating Protein Ensemblescited by 110
- 2025Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matchingcited by 85
- 2023Two for One: Diffusion Models and Force Fields for Coarse-Grained Molecular Dynamicscited by 72
- 2018Machine Learning of coarse-grained Molecular Dynamics Force Fieldscited by 69
- 2020Coarse Graining Molecular Dynamics with Graph Neural Networkscited by 56
- 2024Transferable Boltzmann Generatorscited by 46
- 2024Protein Conformation Generation via Force-Guided SE(3) Diffusion Modelscited by 45
- 2025Scalable Equilibrium Sampling with Sequential Boltzmann Generatorscited by 38
- 2022Machine Learning Coarse-Grained Potentials of Protein Thermodynamicscited by 37
- 2022Flow-matching -- efficient coarse-graining of molecular dynamics without forcescited by 36
25 more
- 2018Coarse-Graining Auto-Encoders for Molecular Dynamicscited by 30
- 2025No Trick, No Treat: Pursuits and Challenges Towards Simulation-free Training of Neural Samplerscited by 28
- 2021Learning neural network potentials from experimental data via Differentiable Trajectory Reweightingcited by 26
- 2017Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape explorationcited by 24
- 2022Simulate Time-integrated Coarse-grained Molecular Dynamics with Multi-Scale Graph Networkscited by 23
- 2023Navigating protein landscapes with a machine-learned transferable coarse-grained modelcited by 23
- 2025Consistent Sampling and Simulation: Molecular Dynamics with Energy-Based Diffusion Modelscited by 23
- 2025Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functionalcited by 20
- 2021Machine Learning Implicit Solvation for Molecular Dynamicscited by 17
- 2024Neural Thermodynamic Integration: Free Energies from Energy-based Diffusion Modelscited by 14
- 2024chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physicscited by 12
- 2025Amortized Sampling with Transferable Normalizing Flowscited by 12
- 2025Scalable Boltzmann Generators for equilibrium sampling of large-scale materialscited by 11
- 2025Transferable Generative Models Bridge Femtosecond to Nanosecond Time-Step Molecular Dynamicscited by 11
- 2025On scalable and efficient training of diffusion samplerscited by 10
- 2017Reinforced dynamics for enhanced sampling in large atomic and molecular systemscited by 7
- 2025Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Samplingcited by 7
- 2025chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulationscited by 6
- 2024AMARO: All Heavy-Atom Transferable Neural Network Potentials of Protein Thermodynamicscited by 3
- 2024Thermodynamic Transferability in Coarse-Grained Force Fields using Graph Neural Networkscited by 2
- 2024Flow Matching for Optimal Reaction Coordinates of Biomolecular Systemcited by 2
- 2025Enhancing Machine Learning Potentials through Transfer Learning across Chemical Elementscited by 2
- 2025Energy-Based Coarse-Graining in Molecular Dynamics: A Flow-Based Framework without Datacited by 2
- 2025Morphology-Aware Peptide Discovery via Masked Conditional Generative Modelingcited by 2
- 2025AI-based Methods for Simulating, Sampling, and Predicting Protein Ensemblescited by 1
Built on by
Citing papers per year: 2026: 3.
3 papers.