Abstract:Here we present a multiscale method to calculate the saddle point associated with the effective dynamics arising from a stochastic system which couples slow deterministic drift and fast stochastic dynamics. This problem is motivated by the transition states on free energy surfaces in chemical physics. Our method is based on the gentlest ascent dynamics which couples the position variable and the direction variable and has the local convergence to saddle points. The dynamics of the direction vector is derived in terms of the covariance function with respective to the equilibrium distribution of the fast stochastic process. We apply the multiscale numerical methods to efficiently solve the obtained multiscale gentlest ascent dynamics, {and discuss the acceleration techniques based on the adaptive idea.} The examples of stochastic ordinary and partial differential equations are presented.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to calculate the saddle point of the effective dynamics in the slow - fast system. Specifically, the author focuses on the saddle point problem of the effective dynamics arising from a stochastic system, which combines slow deterministic drift and fast stochastic dynamics. Such problems have an important application background in the study of transition states on the free - energy surface in chemical physics. ### Main Objectives of the Paper 1. **Calculation of Saddle Points**: The paper aims to develop a multiscale method to calculate the saddle points of the effective dynamics associated with the slow - fast system. Saddle points are of great significance in the study of phase - space structures, especially when used for rate calculations in transition - state theory. 2. **Application of Gentlest Ascent Dynamics (GAD)**: The author adopts the Gentlest Ascent Dynamics method, which can converge locally to saddle points by coupling position variables and direction variables. Gentlest Ascent Dynamics is particularly suitable for effective - dynamics equations without an analytical form. 3. **Multiscale Numerical Methods**: In order to efficiently solve the resulting multiscale Gentlest Ascent Dynamics equations, the author proposes an acceleration technique based on the adaptive idea and discusses related multiscale numerical methods. ### Key Challenges - **Lack of Analytical Form**: Since the effective dynamics \( F \) has no closed form, traditional Newton - type methods cannot be directly applied. - **Complex Time Scales**: There are two different time scales (slow variable \( X \) and fast variable \( Y \)) in the system, and multiscale problems need to be dealt with. - **Impact of Randomness**: The influence of the randomness of the fast variable \( Y \) on the slow variable \( X \) needs to be processed by effective numerical methods. ### Method Overview 1. **Formulation of Gentlest Ascent Dynamics**: Apply Gentlest Ascent Dynamics to the effective dynamics \( \dot{\bar{X}}=F(\bar{X}) \) of the slow - fast system and derive a new multiscale dynamic system. 2. **Multiscale Numerical Methods**: Apply the Heterogeneous Multiscale Method (HMM) or the Seamless Coupling Method (SCM) to efficiently solve the new multiscale system. 3. **Acceleration Techniques**: Introduce adaptive strategies to improve computational efficiency, for example, first use the Seamless Coupling Method to quickly search for the solution area, and then switch to the Heterogeneous Multiscale Method to reduce variance. Through these methods, the author hopes to be able to effectively calculate the saddle points in the effective dynamics of the slow - fast system, thereby providing a powerful tool for transition - state theory and the study of phase - space structures in complex systems.

Multiscale Gentlest Ascent Dynamics for Saddle Point in Effective Dynamics of Slow-Fast System

A Novel Fast Dynamic Optimization Approach for Complex Multivariable Chemical Process Systems

Data-driven Effective Modeling of Multiscale Stochastic Dynamical Systems

The Gentlest Ascent Dynamics

Gentlest Ascent Dynamics on Manifolds Defined by Adaptively Sampled Point-Clouds

Strong convergence of multiscale truncated Euler-Maruyama method for super-linear slow-fast stochastic differential equations

Effective dynamics for a kinetic Monte-Carlo model with slow and fast time scales

An SDE Perspective on Stochastic Inertial Gradient Dynamics with Time-Dependent Viscosity and Geometric Damping

Learning Effective Dynamics from Data-Driven Stochastic Systems

Adaptive Stochastic Gradient Langevin Dynamics: Taming Convergence and Saddle Point Escape Time

A quantitative study on the approximation error and speed-up of the multi-scale MCMC (Monte Carlo Markov chain) method for molecular dynamics

Analysis of Numerical Techniques for Multiscale Stochastic Dynamical Systems

Slow Manifolds for Stochastic Systems with Non-Gaussian Stable Lévy Noise

Non-equilibrium steady states as saddle points and EDP-convergence for slow-fast gradient systems

Atomistic Simulations Of Rare Events Using Gentlest Ascent Dynamics

Slow Foliation of A Slow-Fast Stochastic Evolutionary System

Mathematical and Numerical Analysis to Shrinking-Dimer Saddle Dynamics with Local Lipschitz Conditions

Slow Manifolds for Multi-Time-Scale Stochastic Evolutionary Systems

Analysis of multiscale methods for stochastic dynamical systems driven by α-stable processes

Analysis of Multiscale Methods for Stochastic Differential Equations

On fundamental unifying concepts for trajectory-based slow invariant attracting manifold computation in multiscale models of chemical kinetics