NeSTR: Neural S-Transform Reconstruction for Gaussian SPDEs

DOI10.48550/arXiv.2609.23001MaRDI QIDQ7353455arXivFDO

Authors Fred Espen Benth, Nacira Agram, Jan Rems

Publication date 19 September 2026

Published in arXiv

Tldr We introduce Neural $S$-Transform Reconstruction (NeSTR) for Gaussian SPDEs by approximating their solution's $S$-Transform on a finite dimensional subspace of the white noise test space with neural networks. NeSTR is novel in combining inverse transform reconstruction principles with finite mode learning to provide stable access to higher order chaos coefficients and reusable basis functions.

Full work available at URL https://doi.org/10.48550/arXiv.2609.23001

arXiv classification

math.NAmath.PR

Cites work

• A characterization of Hida distributions
• A review of white noise analysis from a probabilistic standpoint
• Approximation by superpositions of a sigmoidal function
• Approximation theory and approximation practice
• Brownian Motion
• Deep learning based numerical approximation algorithms for stochastic partial differential equations
• Deep operator BSDE: a numerical scheme to approximate solution operators
• Deep Operator BSDE: a Numerical Scheme to Approximate the Solution Operators
• Deep splitting method for parabolic PDEs
• Designing Universal Causal Deep Learning Models: The Case of Infinite-Dimensional Dynamical Systems from Stochastic Analysis
• Fourier Neural Operator for Parametric Partial Differential Equations
• Global universal approximation of functional input maps on weighted spaces
• Introduction to stochastic partial differential equations
• Learning in Modal Space: Solving Time-Dependent Stochastic PDEs Using Physics-Informed Neural Networks
• Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
• Multiple Wiener integral
• Neural Networks in Fr\'echet spaces
• NeuralChaos: Optimal Adapted Approximation of Square Integrable Predictable Processes
• On Distributions of Certain Wiener Functionals
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Pricing options on flow forwards by neural networks in a Hilbert space
• Solving high-dimensional partial differential equations using deep learning
• The orthogonal development of non-linear functionals in series of Fourier-Hermite functionals
• Weighted universal approximation of differentiable maps on infinite-dimensional manifolds

This page was built for publication: NeSTR: Neural S-Transform Reconstruction for Gaussian SPDEs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7353455)