mlsanityData to 7 Oct 2026

Parameterized Representations via Implicit Stochastic Modulation for High-Dimensional and High-Order Neural PDE Solvers

arXiv 2606.22150 ↗First posted 20 Jun 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Solving high-dimensional and high-order PDEs is challenged by the coupled growth of spatial dimensionality and derivative order. Recent stochastic derivative estimators reduce this cost by replacing full derivative tensors with randomized dimension or Taylor estimators, but they are mostly designed for fixed physical parameters and require retraining for each new parameter. We show that direct conditional parameterization of such solvers entangles physical parameters with the high-order automatic differentiation graph, causing extra memory growth and parameter-induced variance amplification. We propose Parameterized Representations via Implicit Stochastic Modulation (PRISM), a plug-and-play framework for parameterized high-dimensional and high-order stochastic neural PDE solvers. PRISM uses a hyper-generator to map physical parameters to affine modulators that scale and shift a purely spatial latent manifold, while keeping parameter branches value-connected but spatial-tangent-disconnected. This design preserves unbiased stochastic dimension and Taylor estimators, removes the parameter encoder from high-order spatial AD, and provides a variance-aware Lipschitz envelope over the parameter space. We prove parameterized unbiasedness, estimation-error bounds, and convergence under bounded stochastic variance. Experiments with PRISM-STDE and PRISM-SDGD on nonlinear parameterized PDEs show stable zero-shot generalization, reduced memory usage, and scalability up to 100,000 dimensions on a single GPU, with efficient low-rank SVD adaptation for unseen parameters.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 56 references that are in the map. First posted on arXiv 20 Jun 2026.

Builds on

33 of 56 references.

References · 33

  1. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  2. 2019DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operatorscited by 1,965
  3. 2017DGM: A deep learning algorithm for solving partial differential equationscited by 674
  4. 2021Characterizing possible failure modes in physics-informed neural networkscited by 632
  5. 2017The Deep Ritz method: A deep learning-based numerical algorithm for solving variational problemscited by 581
  6. 2017Solving high-dimensional partial differential equations using deep learningcited by 549
  7. 2021Physics-Informed Neural Operator for Learning Partial Differential Equationscited by 516
  8. 2021Physics-informed neural networks (PINNs) for fluid mechanics: A reviewcited by 470
  9. 2017Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equationscited by 208
  10. 2017Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equationscited by 188
  11. 2018Adversarial Uncertainty Quantification in Physics-Informed Neural Networkscited by 120
  12. 2020Estimates on the generalization error of Physics Informed Neural Networks (PINNs) for approximating PDEscited by 115
21 more
  1. 2020On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problemscited by 79
  2. 2019Convergence rates for the stochastic gradient descent method for non-convex objective functionscited by 66
  3. 2017Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equationscited by 62
  4. 2021When Do Extended Physics-Informed Neural Networks (XPINNs) Improve Generalization?cited by 57
  5. 2018Forward-Backward Stochastic Neural Networks: Deep Learning of High-dimensional Partial Differential Equationscited by 54
  6. 2019Deep splitting method for parabolic PDEscited by 53
  7. 2019Stochastic Gradient Descent for Nonconvex Learning without Bounded Gradient Assumptionscited by 48
  8. 2022Augmented Physics-Informed Neural Networks (APINNs): A gating network-based soft domain decomposition methodologycited by 43
  9. 2020Two-Layer Neural Networks for Partial Differential Equations: Optimization and Generalization Theorycited by 41
  10. 2021A novel meta-learning initialization method for physics-informed neural networkscited by 35
  11. 2023Learning stiff chemical kinetics using extended deep neural operatorscited by 30
  12. 2018Machine Learning for semi linear PDEscited by 28
  13. 2022Learning Physics-Informed Neural Networks without Stacked Back-propagationcited by 23
  14. 2023Hutchinson Trace Estimation for High-Dimensional and High-Order Physics-Informed Neural Networkscited by 23
  15. 2019Solving high-dimensional optimal stopping problems using deep learningcited by 21
  16. 2024Stochastic Taylor Derivative Estimator: Efficient amortization for arbitrary differential operatorscited by 20
  17. 2023Deep neural operators can serve as accurate surrogates for shape optimization: A case study for airfoilscited by 19
  18. 2021A Priori Generalization Analysis of the Deep Ritz Method for Solving High Dimensional Elliptic Equationscited by 18
  19. 2019Three algorithms for solving high-dimensional fully-coupled FBSDEs through deep learningcited by 15
  20. 2022SVD-PINNs: Transfer Learning of Physics-Informed Neural Networks via Singular Value Decompositioncited by 13
  21. 2022Separable PINN: Mitigating the Curse of Dimensionality in Physics-Informed Neural Networkscited by 5

Built on by

Not cited yet.