Generative methods for sampling transition paths in molecular dynamics
From MaRDI portal
Abstract: Molecular systems often remain trapped for long times around some local minimum of the potential energy function, before switching to another one -- a behavior known as metastability. Simulating transition paths linking one metastable state to another one is difficult by direct numerical methods. In view of the promises of machine learning techniques, we explore in this work two approaches to more efficiently generate transition paths: sampling methods based on generative models such as variational autoencoders, and importance sampling methods based on reinforcement learning.
Cites work
- \({\mathcal Q}\)-learning
- Adaptive Multilevel Splitting for Rare Event Analysis
- Adaptive multilevel splitting in molecular dynamics simulations
- Adaptive multilevel splitting: historical perspective and recent results
- An Introduction to Variational Autoencoders
- Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise.
- Exit probabilities and optimal stochastic control
- Free energy computations. A mathematical perspective
- scientific article; zbMATH DE number 6789886 (Why is no real title available?)
- Introduction to rare event simulation.
- Optimal control of molecular dynamics using Markov state models
- Partial differential equations and stochastic methods in molecular dynamics
- Path sampling with stochastic dynamics: some new algorithms
- Recurrence and invariant measures for degenerate diffusions
- Reinforcement learning. An introduction
- Second-order discretization schemes of stochastic differential systems for the computation of the invariant law
- Transition Path Theory for Markov Jump Processes
Cited in
(3)
This page was built for publication: Generative methods for sampling transition paths in molecular dynamics
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6127065)
