mlsanityData to 7 Oct 2026

chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physics

arXiv 2408.15852 ↗First posted 28 Aug 2024physics.chem-ph cs.LG physics.comp-phcited by 12

Not on the map. Why?

Abstract (arXiv)

Neural Networks (NNs) are effective models for refining the accuracy of molecular dynamics, opening up new fields of application. Typically trained bottom-up, atomistic NN potential models can reach first-principle accuracy, while coarse-grained implicit solvent NN potentials surpass classical continuum solvent models. However, overcoming the limitations of costly generation of accurate reference data and data inefficiency of common bottom-up training demands efficient incorporation of data from many sources. This paper introduces the framework chemtrain to learn sophisticated NN potential models through customizable training routines and advanced training algorithms. These routines can combine multiple top-down and bottom-up algorithms, e.g., to incorporate both experimental and simulation data or pre-train potentials with less costly algorithms. chemtrain provides an object-oriented high-level interface to simplify the creation of custom routines. On the lower level, chemtrain relies on JAX to compute gradients and scale the computations to use available resources. We demonstrate the simplicity and importance of combining multiple algorithms in the examples of parametrizing an all-atomistic model of titanium and a coarse-grained implicit solvent model of alanine dipeptide.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 22 of its 112 references that are in the map. First posted on arXiv 28 Aug 2024.

Builds on

22 of 112 references.

References · 22

  1. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  2. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 701
  3. 2016TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systemscited by 660
  4. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 569
  5. 2020Directional Message Passing for Molecular Graphscited by 556
  6. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  7. 2022Learning Local Equivariant Representations for Large-Scale Atomistic Dynamicscited by 254
  8. 2021Gradients are Not All You Needcited by 82
  9. 2017DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamicscited by 79
  10. 2018Machine Learning of coarse-grained Molecular Dynamics Force Fieldscited by 68
  11. 2020TorchMD: A deep learning framework for molecular simulationscited by 55
  12. 2022Flow-matching -- efficient coarse-graining of molecular dynamics without forcescited by 36
10 more
  1. 2018Coarse-Graining Auto-Encoders for Molecular Dynamicscited by 30
  2. 2022SchNetPack 2.0: A neural network toolbox for atomistic machine learningcited by 29
  3. 2021Learning neural network potentials from experimental data via Differentiable Trajectory Reweightingcited by 26
  4. 2021Machine Learning Implicit Solvation for Molecular Dynamicscited by 16
  5. 2023Accurate machine learning force fields via experimental and simulation data fusioncited by 12
  6. 2022Scalable Bayesian Uncertainty Quantification for Neural Network Potentials: Promise and Pitfallscited by 11
  7. 2024Predicting solvation free energies with an implicit solvent machine learning potentialcited by 10
  8. 2024FeNNol: an Efficient and Flexible Library for Building Force-field-enhanced Neural Network Potentialscited by 6
  9. 2023Top-down machine learning of coarse-grained protein force-fieldscited by 4
  10. 2020Uncertainty estimation for molecular dynamics and samplingcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 3.

12 papers.

  1. 2025chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulationscited by 6
  2. 2025Learning Non-Local Molecular Interactions via Equivariant Local Representations and Charge Equilibrationcited by 5
  3. 2025Understanding and Mitigating Distribution Shifts For Machine Learning Force Fieldscited by 3
  4. 2025Enhanced Sampling for Efficient Learning of Coarse-Grained Machine Learning Potentialscited by 3
  5. 2025Enhancing Machine Learning Potentials through Transfer Learning across Chemical Elementscited by 2
  6. 2025Energy-Based Coarse-Graining in Molecular Dynamics: A Flow-Based Framework without Datacited by 2
  7. 2025Generalization of Long-Range Machine Learning Potentials in Complex Chemical Spacescited by 2
  8. 2025Refining Coarse-Grained Molecular Topologies: A Bayesian Optimization Approachcited by 1
  9. 2025Refining Machine Learning Potentials through Thermodynamic Theory of Phase Transitionscited by 1
  10. 2026Bridging the Simulation-to-Experiment Gap with Generative Models using Adversarial Distribution Alignmentcited by 1
  11. 2026ConSolv: Solvent-Conditional Machine Learning Implicit Solvent Potentialcited by 0
  12. 2026Transferable Implicit Solvent Machine Learning Potential for Drugs and Proteins Approaching Ab Initio Accuracycited by 0