Rectified deep neural networks overcome the curse of dimensionality for nonsmooth value functions in zero-sum games of nonlinear stiff systems

DOI10.1142/S0219530520500116zbMath1456.82804arXiv1903.06652OpenAlexW3031715235MaRDI QIDQ5132232

Yu-Fei Zhang, Christoph Reisinger

Publication date: 10 November 2020

Published in: Analysis and Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1903.06652

zbMATH Keywords

optimal controlapproximation theorystochastic partial differential equationcurse of dimensionalitydeep neural networks

Mathematics Subject Classification ID

Neural nets applied to problems in time-dependent statistical mechanics (82C32) PDEs with randomness, stochastic partial differential equations (35R60) Rate of convergence, degree of approximation (41A25) Finite difference methods applied to problems in statistical mechanics (82M20)

Related Items (20)

Analysis of the Generalization Error: Empirical Risk Minimization over Deep Artificial Neural Networks Overcomes the Curse of Dimensionality in the Numerical Approximation of Black--Scholes Partial Differential Equations ⋮ A Proof that Artificial Neural Networks Overcome the Curse of Dimensionality in the Numerical Approximation of Black–Scholes Partial Differential Equations ⋮ SympOCnet: Solving Optimal Control Problems with Applications to High-Dimensional Multiagent Path Planning Problems ⋮ Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms ⋮ Overcoming the curse of dimensionality in the numerical approximation of parabolic partial differential equations with gradient-dependent nonlinearities ⋮ Full error analysis for the training of deep neural networks ⋮ Unbiased Deep Solvers for Linear Parametric PDEs ⋮ Deep learning methods for partial differential equations and related parameter identification problems ⋮ Numerical methods for backward stochastic differential equations: a survey ⋮ Deep xVA Solver: A Neural Network–Based Counterparty Credit Risk Management Framework ⋮ Neural network approximation and estimation of classifiers with classification boundary in a Barron class ⋮ Solving Kolmogorov PDEs without the curse of dimensionality via deep learning and asymptotic expansion with Malliavin calculus ⋮ An overview on deep learning-based approximation methods for partial differential equations ⋮ Deep ReLU network expression rates for option prices in high-dimensional, exponential Lévy models ⋮ Computing Lyapunov functions using deep neural networks ⋮ Tensor Decomposition Methods for High-dimensional Hamilton--Jacobi--Bellman Equations ⋮ Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning ⋮ Approximate value adjustments for European claims ⋮ DNN expression rate analysis of high-dimensional PDEs: application to option pricing ⋮ A theoretical analysis of deep neural networks and parametric PDEs

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Optimal stochastic control, stochastic target problems, and backward SDE.
• Strong convergence of an explicit numerical method for SDEs with nonglobally Lipschitz continuous coefficients
• Rate of convergence of space time approximations for stochastic evolution equations
• Hybrid switching diffusions. Properties and applications
• Functional analysis, Sobolev spaces and partial differential equations
• Particle representations for a class of nonlinear SPDEs
• Multilayer feedforward networks are universal approximators
• Strong convergence and stability of implicit numerical methods for stochastic differential equations with non-globally Lipschitz continuous coefficients
• Continuous dependence estimates for viscosity solutions of integro-PDEs
• Exponential convergence of the deep neural network approximation for analytic functions
• Optimal approximation of piecewise smooth functions using deep ReLU neural networks
• Error bounds for approximations with deep ReLU networks
• Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
• Deep vs. shallow networks: An approximation theory perspective
• A Stochastic Galerkin Method for Hamilton--Jacobi Equations with Uncertainty
• Error estimates for the finite element discretization of semi-infinite elliptic optimal control problems
• Sparse tensor discretizations of high-dimensional parametric and stochastic PDEs
• Error estimates for the finite element approximation of a semilinear elliptic control problem with state constraints and finite dimensional control space
• Polynomial Approximation of High-Dimensional Hamilton--Jacobi--Bellman Equations and Applications to Feedback Control of Semilinear Parabolic PDEs
• Deep learning in high dimension: Neural network expression rates for generalized polynomial chaos expansions in UQ
• Simulation of stochastic partial differential equations using finite element methods
• Approximation of the Zakai Equation for Nonlinear Filtering
• Stochastic Finite Differences and Multilevel Monte Carlo for a Class of SPDEs in Finance
• Deep backward schemes for high-dimensional nonlinear PDEs
• Control of Systems With Uncertain Initial Conditions
• Optimal Approximation with Sparsely Connected Deep Neural Networks
• New Error Bounds for Deep ReLU Networks Using Sparse Grids
• Sparse grids
• Stochastic Evolution Equations in Portfolio Credit Modelling

This page was built for publication: Rectified deep neural networks overcome the curse of dimensionality for nonsmooth value functions in zero-sum games of nonlinear stiff systems