Deep backward schemes for high-dimensional nonlinear PDEs
From MaRDI portal
Publication:4960067
DOI10.1090/MCOM/3514zbMATH Open1440.60063arXiv1902.01599OpenAlexW2993551995WikidataQ114094322 ScholiaQ114094322MaRDI QIDQ4960067FDOQ4960067
Xavier Warin, Côme Huré, Huyên Pham
Publication date: 8 April 2020
Published in: Mathematics of Computation (Search for Journal in Brave)
Abstract: We propose new machine learning schemes for solving high dimensional nonlinear partial differential equations (PDEs). Relying on the classical backward stochastic differential equation (BSDE) representation of PDEs, our algorithms estimate simultaneously the solution and its gradient by deep neural networks. These approximations are performed at each time step from the minimization of loss functions defined recursively by backward induction. The methodology is extended to variational inequalities arising in optimal stopping problems. We analyze the convergence of the deep learning schemes and provide error estimates in terms of the universal approximation of neural networks. Numerical results show that our algorithms give very good results till dimension 50 (and certainly above), for both PDEs and variational inequalities problems. For the PDEs resolution, our results are very similar to those obtained by the recent method in cite{weinan2017deep} when the latter converges to the right solution or does not diverge. Numerical tests indicate that the proposed methods are not stuck in poor local minimaas it can be the case with the algorithm designed in cite{weinan2017deep}, and no divergence is experienced. The only limitation seems to be due to the inability of the considered deep neural networks to represent a solution with a too complex structure in high dimension.
Full work available at URL: https://arxiv.org/abs/1902.01599
Recommendations
- Approximation error analysis of some deep backward schemes for nonlinear PDEs
- Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning
- DGM: a deep learning algorithm for solving partial differential equations
backward stochastic differential equationsoptimal stopping problemdeep neural networksnonlinear PDEs in high dimension
Probabilistic models, generic numerical methods in probability and statistics (65C20) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Stability and convergence of numerical methods for initial value and initial-boundary value problems involving PDEs (65M12)
Cites Work
- DGM: a deep learning algorithm for solving partial differential equations
- Reflected solutions of backward SDE's, and related obstacle problems for PDE's
- A numerical scheme for BSDEs
- A regression-based Monte Carlo method to solve backward stochastic differential equations
- Adapted solution of a backward stochastic differential equation
- Variational inequalities and the pricing of American options
- Rate of convergence of an empirical regression method for solving generalized backward stochastic differential equations
- Discrete-time approximation and Monte-Carlo simulation of backward stochastic differential equations
- Multilayer feedforward networks are universal approximators
- On irregular functionals of SDEs and the Euler scheme
- Error analysis of the optimal quantization algorithm for obstacle problems.
- Discrete-time approximation for continuously and discretely reflected BSDEs
- On multilevel Picard numerical approximations for high-dimensional nonlinear parabolic partial differential equations and high-dimensional nonlinear backward stochastic differential equations
- Branching diffusion representation of semilinear PDEs and Monte Carlo approximation
- Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
- Title not available (Why is that?)
- Solving high-dimensional partial differential equations using deep learning
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
- Machine learning for semi linear PDEs
- Monte-Carlo Valuation of American Options: Facts and New Algorithms to Improve Existing Methods
- Deep optimal stopping
- Nesting Monte Carlo for high-dimensional non-linear PDEs
- Convergence of the deep BSDE method for coupled FBSDEs
Cited In (83)
- A gradient method for high-dimensional BSDEs
- Convergence of the Backward Deep BSDE Method with Applications to Optimal Stopping Problems
- Lax-Oleinik-type formulas and efficient algorithms for certain high-dimensional optimal control problems
- The Calderón's problem via DeepONets
- Efficient pricing and hedging of high-dimensional American options using deep recurrent networks
- Deep learning scheme for forward utilities using ergodic BSDEs
- High-dimensional stochastic control models for newsvendor problems and deep learning resolution
- A neural network approach for stochastic optimal control
- Numerical approximations of coupled forward–backward SPDEs
- Learning the random variables in Monte Carlo simulations with stochastic gradient descent: Machine learning for parametric PDEs and financial derivative pricing
- Convergence of a Robust Deep FBSDE Method for Stochastic Control
- Learning High-Dimensional McKean–Vlasov Forward-Backward Stochastic Differential Equations with General Distribution Dependence
- Recent developments in machine learning methods for stochastic control and games
- A Neural Network Approach to High-Dimensional Optimal Switching Problems with Jumps in Energy Markets
- Numerical solution of the modified and non-Newtonian Burgers equations by stochastic coded trees
- Deep Curve-Dependent PDEs for Affine Rough Volatility
- Neural Control of Parametric Solutions for High-Dimensional Evolution PDEs
- A deep branching solver for fully nonlinear partial differential equations
- Pathwise CVA regressions with oversimulated defaults
- Achieving high convergence rates by quasi-Monte Carlo and importance sampling for unbounded integrands
- Deep learning algorithms for solving high-dimensional nonlinear backward stochastic differential equations
- Deep learning approximations for non-local nonlinear PDEs with Neumann boundary conditions
- Neural networks for first order HJB equations and application to front propagation with obstacle terms
- Deep signature algorithm for multidimensional path-dependent options
- Efficient and stable SAV-based methods for gradient flows arising from deep learning
- Numerical approximation of Dynkin games with asymmetric information
- The Kolmogorov infinite dimensional equation in a Hilbert space via deep learning methods
- Relative wealth concerns with partial information and heterogeneous priors
- Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space
- Deep learning methods for partial differential equations and related parameter identification problems
- A deep learning approach to the probabilistic numerical solution of path-dependent partial differential equations
- McKean Feynman-Kac probabilistic representations of non-linear partial differential equations
- Deep neural network framework based on backward stochastic differential equations for pricing and hedging American options in high dimensions
- Solving stochastic optimal control problem via stochastic maximum principle with deep learning method
- Overcoming the curse of dimensionality in the numerical approximation of parabolic partial differential equations with gradient-dependent nonlinearities
- Rectified deep neural networks overcome the curse of dimensionality for nonsmooth value functions in zero-sum games of nonlinear stiff systems
- Approximate value adjustments for European claims
- An efficient weak Euler-Maruyama type approximation scheme of very high dimensional SDEs by orthogonal random variables
- A fully nonlinear Feynman-Kac formula with derivatives of arbitrary orders
- Reinforcement learning and stochastic optimisation
- An overview on deep learning-based approximation methods for partial differential equations
- Overcoming the curse of dimensionality in the numerical approximation of Allen-Cahn partial differential equations via truncated full-history recursive multilevel Picard approximations
- Linear Convergence of a Policy Gradient Method for Some Finite Horizon Continuous Time Control Problems
- Deep learning schemes for parabolic nonlocal integro-differential equations
- Pricing Options under Rough Volatility with Backward SPDEs
- Deep neural networks based temporal-difference methods for high-dimensional parabolic partial differential equations
- Overcoming the curse of dimensionality in the approximative pricing of financial derivatives with default risks
- APFOS-Net: asymptotic preserving scheme for anisotropic elliptic equations with deep neural network
- Neural networks-based backward scheme for fully nonlinear PDEs
- Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms
- On quadrature rules for solving partial differential equations using neural networks
- Gradient boosting-based numerical methods for high-dimensional backward stochastic differential equations
- Discretization and machine learning approximation of BSDEs with a constraint on the gains-process
- Computing Lyapunov functions using deep neural networks
- XVA analysis from the balance sheet
- Solving non-linear Kolmogorov equations in large dimensions by using deep learning: a numerical comparison of discretization schemes
- Deep Splitting Method for Parabolic PDEs
- The deep parametric PDE method and applications to option pricing
- A level-set approach for stochastic optimal control problems under controlled-loss constraints
- Numerical resolution of McKean-Vlasov FBSDEs using neural networks
- Unbiased Deep Solvers for Linear Parametric PDEs
- Convergence of deep fictitious play for stochastic differential games
- Title not available (Why is that?)
- Tensor Decomposition Methods for High-dimensional Hamilton--Jacobi--Bellman Equations
- Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning
- Control variate method for deep BSDE solver using weak approximation
- DeepSets and their derivative networks for solving symmetric PDEs
- Numerical approximation of singular forward-backward SDEs
- Higher-order deep solver of non-linear PDEs implied by a non-linear discrete Clark-Ocone formula
- Learning a functional control for high-frequency finance
- Overcoming the curse of dimensionality for some Hamilton-Jacobi partial differential equations via neural network architectures
- Numerical methods for backward stochastic differential equations: a survey
- Multilevel Picard approximations of high-dimensional semilinear partial differential equations with locally monotone coefficient functions
- Deep Weak Approximation of SDEs: A Spatial Approximation Scheme for Solving Kolmogorov Equations
- Solving Kolmogorov PDEs without the curse of dimensionality via deep learning and asymptotic expansion with Malliavin calculus
- SympOCnet: Solving Optimal Control Problems with Applications to High-Dimensional Multiagent Path Planning Problems
- Deep xVA Solver: A Neural Network–Based Counterparty Credit Risk Management Framework
- Stability of backward stochastic differential equations: the general Lipschitz case
- Approximation Error Analysis of Some Deep Backward Schemes for Nonlinear PDEs
- An application of the splitting-up method for the computation of a neural network representation for the solution for the filtering equations
- Deep combinatorial optimisation for optimal stopping time problems: application to swing options pricing.
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Extensions of the deep Galerkin method
Uses Software
This page was built for publication: Deep backward schemes for high-dimensional nonlinear PDEs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4960067)
