Optimal control for sampling the transition path process and estimating rates
From MaRDI portal
Abstract: Many processes in nature such as conformal changes in biomolecules and clusters of interacting particles, genetic switches, noisy mechanical or electromechanical oscillators, and many others are modeled using stochastic differential equations with small white noise. The study of rare transitions between metastable states in such systems is of great interest and importance, but direct simulations are difficult due to long waiting times. Transition path theory is a mathematical framework for the quantitative description of rare events. Its direct implementation the key component of which is the solution of the committor problem, a boundary value problem for the backward Kolmogorov equation, is often challenging due to high dimensionality or other numerical issues. This work exploits the key fact that the optimal controller constructed from the committor leads to generation of transition trajectories exclusively.
Recommendations
- Transition Path Theory for Langevin Dynamics on Manifolds: Optimal Control and Data-Driven Solver
- scientific article; zbMATH DE number 5689010
- A finite expression method for solving high-dimensional committor problems
- A numerical scheme for optimal transition paths of stochastic chemical kinetic systems
- Data-driven method to learn the most probable transition pathway and stochastic differential equation
Cites work
- A Koopman framework for rare event simulation in stochastic differential equations
- A Simple Mesh Generator in MATLAB
- Adaptive Multilevel Splitting for Rare Event Analysis
- Adaptive multilevel splitting: historical perspective and recent results
- An efficient jet marcher for computing the quasipotential for 2D SDEs. Enhancing accuracy and efficiency of quasipotential solvers
- An ordered line integral method for computing the quasi-potential in the case of variable anisotropic diffusion
- ATLAS: a geometric approach to learning high-dimensional stochastic systems near manifolds
- Computing committors in collective variables via Mahalanobis diffusion maps
- Computing the quasipotential for highly dissipative and chaotic SDEs an application to stochastic Lorenz'63
- Computing the quasipotential for nongradient SDEs in 3D
- Controlled Markov processes and viscosity solutions
- Diffusion maps tailored to arbitrary non-degenerate Itô processes
- Diffusion Maps, Reduction Coordinates, and Low Dimensional Representation of Stochastic Systems
- Effective dynamics using conditional expectations
- Exponential convergence of Langevin distributions and their discrete approximations
- Finding the quasipotential for nongradient SDEs
- Flows in complex networks: theory, algorithms, and application to lennard-Jones cluster rearrangement
- scientific article; zbMATH DE number 3878293 (Why is no real title available?)
- On the Hill relation and the mean reaction time for metastable processes
- Optimal control of molecular dynamics using Markov state models
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Random Perturbations of Dynamical Systems
- Reactive trajectories and the transition path process
- Remarks around 50 lines of Matlab: short finite element implementation
- Some remarks on the Smoluchowski-Kramers approximation
- Stochastic differential equations. An introduction with applications.
- Towards a theory of transition paths
- Transition Path Theory for Langevin Dynamics on Manifolds: Optimal Control and Data-Driven Solver
Cited in
(2)
This page was built for publication: Optimal control for sampling the transition path process and estimating rates
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6189442)
