Machine learning kinetics from molecular dynamics data
Not on the map. Why?
Abstract (arXiv)
Most molecular transitions occur on timescales far beyond direct molecular dynamics simulations. The committor, the probability that a configuration reaches a product state before a reactant state, is a central kinetic statistic, providing a mechanism-independent reaction coordinate and a foundation for transition path theory and the calculation of rates. This review surveys modern approaches for estimating the committor and related kinetic statistics from molecular simulations, with an emphasis on self-supervised methods that learn solutions of their defining dynamical equations rather than relying on labeled shooting data. We develop a common operator viewpoint connecting generator-based partial differential equations, variational principles, Markov state models, dynamical Galerkin approximation, and neural networks. Empirical and theoretical evidence points to the efficiency of these methods. We provide theoretical and practical guidance for realizing their full potential in applications, including strategies for treating non-Markovian effects and for sampling. We conclude by identifying opportunities for further research, including connections to reinforcement learning and generative modeling.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 34 of its 232 references that are in the map. First posted on arXiv 15 Sep 2026.
Builds on
34 of 232 references.
References · 34
- 2017DGM: A deep learning algorithm for solving partial differential equationscited by 678
- 2017The Deep Ritz method: A deep learning-based numerical algorithm for solving variational problemscited by 583
- 2017Solving high-dimensional partial differential equations using deep learningcited by 556
- 2024InterPLM: Discovering Interpretable Features in Protein Language Models via Sparse Autoencoderscited by 66
- 2022Stochastic Optimal Control for Collective Variable Free Sampling of Molecular Transition Pathscited by 38
- 2020Solving high-dimensional eigenvalue problems using deep neural networks: A diffusion Monte Carlo like approachcited by 35
- 2018Solving for high dimensional committor functions using artificial neural networkscited by 34
- 2019Computing Committor Functions for the Study of Rare Events Using Deep Learningcited by 29
- 2024Doob's Lagrangian: A Sample-Efficient Variational Approach to Transition Path Samplingcited by 27
- 2025Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functionalcited by 20
- 2016Data-driven parameterization of the generalized Langevin equationcited by 14
- 2019Variational approach to rare event simulation using least-squares regressioncited by 12
22 more
- 2019High-resolution Markov state models for the dynamics of Trp-cage miniprotein constructed over slow folding modes identified by state-free reversible VAMPnetscited by 11
- 2023Designing losses for data-free training of normalizing flows on Boltzmann distributionscited by 9
- 2023Inexact iterative numerical linear algebra for neural network-based spectral estimation and rare-event predictioncited by 9
- 2023Diffusion Methods for Generating Transition Pathscited by 8
- 2024The surprising efficiency of temporal difference learning for rare event predictioncited by 7
- 2022Generative methods for sampling transition paths in molecular dynamicscited by 6
- 2025Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimationcited by 6
- 2021Supervised Learning and the Finite-Temperature String Method for Computing Committor Functions and Reaction Ratescited by 5
- 2023Optimal control for sampling the transition path process and estimating ratescited by 5
- 2024Deep Learning Method for Computing Committor Functions with Adaptive Samplingcited by 5
- 2024The fast committor machine: Interpretable prediction with kernelscited by 5
- 2024Graph Neural Network-State Predictive Information Bottleneck (GNN-SPIB) approach for learning molecular thermodynamics and kineticscited by 5
- 2024Using pretrained graph neural networks with token mixers as geometric featurizers for conformational dynamicscited by 5
- 2024Deep Learning for Computing Convergence Rates of Markov Chainscited by 4
- 2025Estimating Committor Functions via Deep Adaptive Sampling on Rare Transition Pathscited by 4
- 2026Rare Event Analysis via Stochastic Optimal Controlcited by 4
- 2025Deep Learning for Markov Chains: Lyapunov Functions, Poisson's Equation, and Stationary Distributionscited by 2
- 2022Weighted ensemble: Recent mathematical developmentscited by 1
- 2025An exact multiple-time-step variational formulation for the committor and the transition ratecited by 1
- 2025Hierarchical geometric deep learning enables scalable analysis of molecular dynamicscited by 1
- 2026Quantum statistics from classical simulations via generative Gibbs samplingcited by 1
- 2026Composing diffusion priors with explicit physical context via generative Gibbs samplingcited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2022Prediction of good reaction coordinates and future evolution of MD trajectories using Regularized Sparse Autoencoders: A novel deep learning approach
- 2024Generalized Flow Matching for Transition Dynamics Modeling
- 2023Score dynamics: scaling molecular dynamics with picoseconds timestep via conditional diffusion model
- 2023Implicit Transfer Operator Learning: Multiple Time-Resolution Surrogates for Molecular Dynamics
- 2025Learning Biomolecular Motion: The Physics-Informed Machine Learning Paradigm
- 2026A Priori Sampling of Transition States with Guided Diffusion
- 2026Driving Reaction Trajectories via Latent Flow Matching
- 2025Efficient Exploration of Chemical Kinetics