Solving Fokker-Planck equation using deep learning

Abstract

The probability density function of stochastic differential equations is governed by the Fokker-Planck (FP) equation. A novel machine learning method is developed to solve the general FP equations based on deep neural networks. The proposed algorithm does not require any interpolation and coordinate transformation, which is different from the traditional numercial methods. The main novelty of this paper is that penalty factors are introduced to overcome the local optimization for the deep learning approach, and the corresponding setting rules are given. Meanwhile, we consider a normalization condition as a supervision condition to effectively avoid that the trial solution is zero. Several numerical examples are presented to illustrate performances of the proposed algorithm, including one- and two-dimensional systems. All the results suggest that the deep learning is quite feasible and effective to calculate the FP equation. Further, influences of the number of hidden layers, the penalty factors, and the optimization algorithm are discussed in detail. These results indicate that the performances of the machine learning technique can be improved through constructing the neural networks appropriately.

References in corpus (9)

• Artificial Neural Networks for Solving Ordinary and Partial Differential Equations
• DGM: A deep learning algorithm for solving partial differential equations
• Solving high-dimensional partial differential equations using deep learning
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• A unified deep artificial neural network approach to partial differential equations in complex geometries
• Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations
• Lévy Noise Induced Switch in the Gene Transcriptional Regulatory System
• A discontinuous Galerkin method for nonlinear parabolic equations and gradient flow problems with interaction potentials
• Monte Carlo method and High Performance Computing for solving Fokker-Planck equation of minority plasma particles

Cited by in corpus (17)

• A Machine Learning Framework for Computing the Most Probable Paths of Stochastic Dynamical Systems
• Adversarial Multi-task Learning Enhanced Physics-informed Neural Networks for Solving Partial Differential Equations
• Solution of FPK Equation for Stochastic Dynamics Subjected to Additive Gaussian Noise via Deep Learning Approach
• Data-Driven Statistical Reduced-Order Modeling and Quantification of Polycrystal Mechanics Leading to Porosity-Based Ductile Damage
• Solving Inverse Stochastic Problems from Discrete Particle Observations Using the Fokker-Planck Equation and Physics-informed Neural Networks
• Mitigation of rare events in multistable systems driven by correlated noise
• Solving Non-local Fokker-Planck Equations by Deep Learning
• Dynamical analysis of a parameter-aware reservoir computer
• Detecting Stochastic Governing Laws with Observation on Stationary Distributions
• LSTM-PINN: An Hybrid Method for Prediction of Steady-State Electrohydrodynamic Flow
• Density Planner: Minimizing Collision Risk in Motion Planning with Dynamic Obstacles using Density-based Reachability
• An energy-based deep splitting method for the nonlinear filtering problem
• Solving physics-based initial value problems with unsupervised machine learning
• Optimal Transport Based Refinement of Physics-Informed Neural Networks
• Learning collision operators from plasma phase space data using differentiable simulators
• Physics-Informed Deep Learning Characterizes Morphodynamics of Asian Soybean Rust Disease
• Stationary Density Estimation of Itô Diffusions Using Deep Learning