Neural networks for first order HJB equations and application to front propagation with obstacle terms

DOI10.1007/S42985-023-00258-8MaRDI QIDQ6087416zbMATH OpenOpenAlexFDO

Authors Olivier Bokanowski, Averil Prost, Xavier Warin

Publication date 15 November 2023

Published in SN Partial Differential Equations and Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2210.04300

zbMATH Keywords

neural networksstate constraintsdynamic programming principledeterministic optimal controlfirst order Hamilton-Jacobi-Bellman equation

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Hamilton-Jacobi equations (35F21) Dynamic programming in optimal control and differential games (49L20)

Abstract: We consider a deterministic optimal control problem with a maximum running cost functional, in a finite horizon context, and propose deep neural network approximations for Bellman's dynamic programming principle, corresponding also to some first-order Hamilton-Jacobi-Bellman equations. This work follows the lines of Hur'e et al. (SIAM J. Numer. Anal., vol. 59 (1), 2021, pp. 525-557) where algorithms are proposed in a stochastic context. However, we need to develop a completely new approach in order to deal with the propagation of errors in the deterministic setting, where no diffusion is present in the dynamics. Our analysis gives precise error estimates in an average norm. The study is then illustrated on several academic numerical examples related to front propagations models in the presence of obstacle constraints, showing the relevance of the approach for average dimensions (e.g. from $2$ to $8$), even for non-smooth value functions.

Recommendations

• Approximate solution of the Hamilton-Jacobi-Bellman equation
• scientific article; zbMATH DE number 2015380
• Iterative computational approach to the solution of the Hamilton-Jacobi-Bellman-lsaacs equation in nonlinear optimal control
• The rate of convergence of finite-difference approximations for Bellman equations with Lipschitz coefficients

Cites work

• scientific article; zbMATH DE number 1577097 (Why is no real title available?)
• scientific article; zbMATH DE number 1405762 (Why is no real title available?)
• A Discontinuous Galerkin Finite Element Method for Hamilton--Jacobi Equations
• A discontinuous Galerkin scheme for front propagation with obstacles
• A distribution-free theory of nonparametric regression
• A forward scheme for backward SDEs
• A general Hamilton-Jacobi framework for non-linear state-constrained control problems
• A regression-based Monte Carlo method to solve backward stochastic differential equations
• A tree structure algorithm for optimal control problems with state constraints
• An adaptive sparse grid semi-Lagrangian scheme for first order Hamilton-Jacobi Bellman equations
• Approximation error analysis of some deep backward schemes for nonlinear PDEs
• Convergence of discontinuous Galerkin schemes for front propagation with obstacles
• Convergence of the deep BSDE method for coupled FBSDEs
• DGM: a deep learning algorithm for solving partial differential equations
• Deep backward schemes for high-dimensional nonlinear PDEs
• Deep neural networks algorithms for stochastic control problems on finite horizon: convergence analysis
• Deep neural networks algorithms for stochastic control problems on finite horizon: numerical applications
• Discontinuous Galerkin finite element methods for time-dependent Hamilton-Jacobi-Bellman equations with Cordes coefficients
• Handbook of numerical methods for hyperbolic problems. Basic and fundamental issues
• High-Order Essentially Nonoscillatory Schemes for Hamilton–Jacobi Equations
• Linear regression MDP scheme for discrete backward stochastic differential equations under general conditions
• Numerical methods for the pricing of swing options: a stochastic control approach
• On the Convergence of Finite Element Methods for Hamilton--Jacobi--Bellman Equations
• Optimistic planning algorithms for state-constrained optimal control problems
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Reachability and Minimal Times for State Constrained Nonlinear Problems without Any Controllability Assumption
• Reinterpretation and simplified implementation of a discontinuous Galerkin method for Hamilton-Jacobi equations
• Solving high-dimensional partial differential equations using deep learning
• Suboptimal feedback control of PDEs by solving HJB equations on adaptive sparse grids
• Tensor decomposition methods for high-dimensional Hamilton-Jacobi-Bellman equations
• The Max-Plus Finite Element Method for Solving Deterministic Optimal Control Problems: Basic Properties and Convergence Analysis
• Two Approximations of Solutions of Hamilton-Jacobi Equations
• Value function and optimal trajectories for a maximum running cost control problem with state constraints. Application to an abort landing problem

Cited in

(1)

• A multilevel fast marching method for the minimum time problem

This page was built for publication: Neural networks for first order HJB equations and application to front propagation with obstacle terms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6087416)