Computing the Invariant Distribution of Randomly Perturbed Dynamical
Systems Using Deep Learning
- URL: http://arxiv.org/abs/2110.11538v1
- Date: Fri, 22 Oct 2021 00:45:46 GMT
- Title: Computing the Invariant Distribution of Randomly Perturbed Dynamical
Systems Using Deep Learning
- Authors: Bo Lin, Qianxiao Li, Weiqing Ren
- Abstract summary: Invariant distribution is an important object in the study of randomly perturbed dynamical systems.
Traditional numerical methods for computing the invariant distribution based on the Fokker-Planck equation are limited to low-dimensional systems.
We propose a deep learning based method to compute the generalized potential.
- Score: 9.053926666240118
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: The invariant distribution, which is characterized by the stationary
Fokker-Planck equation, is an important object in the study of randomly
perturbed dynamical systems. Traditional numerical methods for computing the
invariant distribution based on the Fokker-Planck equation, such as finite
difference or finite element methods, are limited to low-dimensional systems
due to the curse of dimensionality. In this work, we propose a deep learning
based method to compute the generalized potential, i.e. the negative logarithm
of the invariant distribution multiplied by the noise. The idea of the method
is to learn a decomposition of the force field, as specified by the
Fokker-Planck equation, from the trajectory data. The potential component of
the decomposition gives the generalized potential. The method can deal with
high-dimensional systems, possibly with partially known dynamics. Using the
generalized potential also allows us to deal with systems at low temperatures,
where the invariant distribution becomes singular around the metastable states.
These advantages make it an efficient method to analyze invariant distributions
for practical dynamical systems. The effectiveness of the proposed method is
demonstrated by numerical examples.
Related papers
- Learning Controlled Stochastic Differential Equations [61.82896036131116]
This work proposes a novel method for estimating both drift and diffusion coefficients of continuous, multidimensional, nonlinear controlled differential equations with non-uniform diffusion.
We provide strong theoretical guarantees, including finite-sample bounds for (L2), (Linfty), and risk metrics, with learning rates adaptive to coefficients' regularity.
Our method is available as an open-source Python library.
arXiv Detail & Related papers (2024-11-04T11:09:58Z) - Total Uncertainty Quantification in Inverse PDE Solutions Obtained with Reduced-Order Deep Learning Surrogate Models [50.90868087591973]
We propose an approximate Bayesian method for quantifying the total uncertainty in inverse PDE solutions obtained with machine learning surrogate models.
We test the proposed framework by comparing it with the iterative ensemble smoother and deep ensembling methods for a non-linear diffusion equation.
arXiv Detail & Related papers (2024-08-20T19:06:02Z) - Weak Collocation Regression for Inferring Stochastic Dynamics with
L\'{e}vy Noise [8.15076267771005]
We propose a weak form of the Fokker-Planck (FP) equation for extracting dynamics with L'evy noise.
Our approach can simultaneously distinguish mixed noise types, even in multi-dimensional problems.
arXiv Detail & Related papers (2024-03-13T06:54:38Z) - Physics-Informed Solution of The Stationary Fokker-Plank Equation for a
Class of Nonlinear Dynamical Systems: An Evaluation Study [0.0]
An exact analytical solution of the Fokker-Planck (FP) equation is only available for a limited subset of dynamical systems.
To evaluate its potential, we present a data-free, physics-informed neural network (PINN) framework to solve the FP equation.
arXiv Detail & Related papers (2023-09-25T13:17:34Z) - Decimation technique for open quantum systems: a case study with
driven-dissipative bosonic chains [62.997667081978825]
Unavoidable coupling of quantum systems to external degrees of freedom leads to dissipative (non-unitary) dynamics.
We introduce a method to deal with these systems based on the calculation of (dissipative) lattice Green's function.
We illustrate the power of this method with several examples of driven-dissipative bosonic chains of increasing complexity.
arXiv Detail & Related papers (2022-02-15T19:00:09Z) - Structure-Preserving Learning Using Gaussian Processes and Variational
Integrators [62.31425348954686]
We propose the combination of a variational integrator for the nominal dynamics of a mechanical system and learning residual dynamics with Gaussian process regression.
We extend our approach to systems with known kinematic constraints and provide formal bounds on the prediction uncertainty.
arXiv Detail & Related papers (2021-12-10T11:09:29Z) - Differential Equation Based Path Integral for System-Bath Dynamics [0.0]
We propose the differential equation based path integral (DEBPI) method to simulate the real-time evolution of open quantum systems.
New numerical schemes can be derived by discretizing these differential equations.
It is numerically verified that in certain cases, by selecting appropriate systems and applying suitable numerical schemes, the memory cost required in the i-QuAPI method can be significantly reduced.
arXiv Detail & Related papers (2021-07-22T15:06:22Z) - Large-Scale Wasserstein Gradient Flows [84.73670288608025]
We introduce a scalable scheme to approximate Wasserstein gradient flows.
Our approach relies on input neural networks (ICNNs) to discretize the JKO steps.
As a result, we can sample from the measure at each step of the gradient diffusion and compute its density.
arXiv Detail & Related papers (2021-06-01T19:21:48Z) - Learning High-Dimensional Distributions with Latent Neural Fokker-Planck
Kernels [67.81799703916563]
We introduce new techniques to formulate the problem as solving Fokker-Planck equation in a lower-dimensional latent space.
Our proposed model consists of latent-distribution morphing, a generator and a parameterized Fokker-Planck kernel function.
arXiv Detail & Related papers (2021-05-10T17:42:01Z) - A Data Driven Method for Computing Quasipotentials [8.055813148141246]
The quasipotential plays a central role in characterizing statistics of transition events and likely transition paths.
Traditional methods based on the dynamic programming principle or path space tend to suffer from the curse of dimensionality.
We show that our method can effectively compute quasipotential landscapes without requiring spatial discretization or solving path-space optimization problems.
arXiv Detail & Related papers (2020-12-13T02:32:49Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.