Energy-Guided Learning for Heat Equation with Noisy Source: Optimal Estimation and Generalization Bounds
Abstract
Data and the governing partial differential equations (PDEs) of continuous-time dynamical systems provide rich structure for statistical learning. A central challenge is how to integrate data with PDE constraints in a principled way to improve generalization. In this paper, we characterize the statistical estimation and generalization properties of learning solutions to the heat equation with a noisy source function from finite, randomly sampled data. We first derive a novel variational formulation, which we characterize as the minimizer of loss function for an energy regularized and ellipticity debiased Sobolev regression. By formulating the learning task as a supervised problem over space-time domains under physically meaningful norms, our approach goes beyond the traditional supervised learning framework by effectively leveraging \emph{auxiliary data}---samples of the source function of the heat equation---to design a \emph{physics-informed energy regularization}. Our analysis reveals that learning in linear parabolic PDEs exhibits intrinsic anisotropy between spatial and temporal regularity, in sharp contrast to the static elliptic problems, leading to fundamentally different estimation and generalization behaviors. Specifically, we derive non-asymptotically rate-optimal estimation errors, as well as generalization error, which explicitly capture the interplay among temporal sampling, spatial resolution, and solution smoothness. Our numerical experiments demonstrate substantial orders of magnitude of reduction in testing errors against various physics-informed neural network methods on various parabolic and hyperbolic PDEs.
Related
- Is Spectral Self-Supervised Learning a Minimax Optimal Method of Density Ratio Estimation?
- Learning under Target Shift: Optimal Density Ratio Estimation and Importance-Weighted Regression
- A Generalization of the Riemann Functional Equation with Three Variables
- Hydrodynamic limits of the asymmetric Brownian energy process and its dynamic generalization
- Optimal guaranteed sequential estimation for mean-reverting CEV diffusions
BibTeX
@misc{zhou2026energyguided,
title = {{Energy-Guided Learning for Heat Equation with Noisy Source: Optimal Estimation and Generalization Bounds}},
author = {Zixiang Zhou and Xiaohui Chen and Shanghua Teng},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.15111},
eprint = {2610.15111},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.15111},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Energy-Guided Learning for Heat Equation with Noisy Source: Optimal Estimation and Generalization Bounds
AU - Zixiang Zhou
AU - Xiaohui Chen
AU - Shanghua Teng
PY - 2026
DA - 2026/10/10
PB - arXiv
DO - 10.48550/arxiv.2610.15111
UR - https://arxiv.org/abs/2610.15111
AB - Data and the governing partial differential equations (PDEs) of continuous-time dynamical systems provide rich structure for statistical learning. A central challenge is how to integrate data with PDE constraints in a principled way to improve generalization. In this paper, we characterize the statistical estimation and generalization properties of learning solutions to the heat equation with a noisy source function from finite, randomly sampled data. We first derive a novel variational formulation, which we characterize as the minimizer of loss function for an energy regularized and ellipticity debiased Sobolev regression. By formulating the learning task as a supervised problem over space-time domains under physically meaningful norms, our approach goes beyond the traditional supervised learning framework by effectively leveraging \emph{auxiliary data}---samples of the source function of the heat equation---to design a \emph{physics-informed energy regularization}. Our analysis reveals that learning in linear parabolic PDEs exhibits intrinsic anisotropy between spatial and temporal regularity, in sharp contrast to the static elliptic problems, leading to fundamentally different estimation and generalization behaviors. Specifically, we derive non-asymptotically rate-optimal estimation errors, as well as generalization error, which explicitly capture the interplay among temporal sampling, spatial resolution, and solution smoothness. Our numerical experiments demonstrate substantial orders of magnitude of reduction in testing errors against various physics-informed neural network methods on various parabolic and hyperbolic PDEs.
ER -