analiza aiqbench
This review bridges computational physics and machine learning by applying self-supervised neural networks to solve dynamical equations for molecular kinetics, effectively bypassing the need for labeled shooting data in simulations. It establishes a unified operator framework connecting partial differential equations, Markov state models, and variational principles to accelerate the analysis of rare events on timescales inaccessible to direct simulation. The work highlights significant efficiency gains while identifying future intersections with reinforcement learning and generative modeling for scientific discovery.
Machine learning kinetics from molecular dynamics data
arXiv:2609.17736v1 Announce Type: cross Abstract: 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.
Abstract: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.
Computational Physics (physics.comp-ph); Statistical Mechanics (cond-mat.stat-mech); Machine Learning (cs.LG); Chemical Physics (physics.chem-ph); Data Analysis, Statistics and Probability (physics.data-an)
arXiv:2609.17736 [physics.comp-ph]
(or arXiv:2609.17736v1 [physics.comp-ph] for this version)
https://doi.org/10.48550/arXiv.2609.17736 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 190 cuvinte
Subiecte conexe
News
Quantum Computing Report · 7 iun. 2026
Langevin-Informed Transfer Learning: Replacing Target Samples by Black-Box Feedback
arXiv · 2 oct. 2026
Learning Topological Representations of Protein Structure and Dynamics
arXiv · 6 oct. 2026
Learning Topological Representations for Molecular Dynamics
arXiv · 16 iun. 2026