Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data

• URL: http://arxiv.org/abs/2510.23111v1
• Date: Mon, 27 Oct 2025 08:31:55 GMT
• Title: Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data
• Abstract summary: We show neural networks trained purely on low-fidelity solver data can achieve higher accuracy than those solvers when evaluated against a higher-fidelity reference.<n>We empirically validate this finding across different PDEs using standard neural architectures.<n>This work prompts a re-evaluation of emulator benchmarking, suggesting neural emulators might achieve greater physical fidelity than their training source within specific operational regimes.
• Score: 28.57551605623958
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural operators or emulators for PDEs trained on data from numerical solvers are conventionally assumed to be limited by their training data's fidelity. We challenge this assumption by identifying "emulator superiority," where neural networks trained purely on low-fidelity solver data can achieve higher accuracy than those solvers when evaluated against a higher-fidelity reference. Our theoretical analysis reveals how the interplay between emulator inductive biases, training objectives, and numerical error characteristics enables superior performance during multi-step rollouts. We empirically validate this finding across different PDEs using standard neural architectures, demonstrating that emulators can implicitly learn dynamics that are more regularized or exhibit more favorable error accumulation properties than their training data, potentially surpassing training data limitations and mitigating numerical artifacts. This work prompts a re-evaluation of emulator benchmarking, suggesting neural emulators might achieve greater physical fidelity than their training source within specific operational regimes. Project Page: https://tum-pbs.github.io/emulator-superiority

Related papers

• From Numerical Simulators of PDEs to Neural Emulators and Back [2.9612444540570113]
  This thesis argues neural architectures mirror classical discretizations, their errors are amenable to the same spectral analysis, and insight flows profitably in both directions.<n>Mode-wise Fourier analysis then provides a common language in which solver errors, architectural inductive biases, and training objectives can all be read off simultaneously.
  arXiv  Detail & Related papers  (2026-08-25T13:34:15Z)
• Improving Long-Range Interactions in Graph Neural Simulators via Hamiltonian Dynamics [71.53370807809296]
  Recent Graph Neural Simulators (GNSs) accelerate simulations by learning dynamics on graph-structured data.<n>We propose Information-preserving Graph Neural Simulators (IGNS), a graph-based neural simulator built on the principles of Hamiltonian dynamics.<n>IGNS consistently outperforms state-of-the-art GNSs, achieving higher accuracy and stability under challenging and complex dynamical systems.
  arXiv  Detail & Related papers  (2025-11-11T12:53:56Z)
• LLMs as Scalable, General-Purpose Simulators For Evolving Digital Agent Training [55.72784274656801]
  We introduce a scalable paradigm that generates structured UI states and transitions to synthesize training trajectories at scale.<n>Our paradigm integrates a digital world simulator for diverse UI states, a guided rollout process for coherent exploration, and a trajectory wrapper.<n>Experiments on WebArena and AndroidWorld show that UI-Simulator rivals or surpasses open-source agents trained on real UIs with significantly better robustness.
  arXiv  Detail & Related papers  (2025-10-16T17:59:38Z)
• Data-Augmented Few-Shot Neural Emulator for Computer-Model System Identification [20.49905192303411]
  Partial equations (PDEs) are the modeling of many natural and engineered systems.<n>It can be convenient to express such models as neural PDEs by replacing part or all of the PDE's governing equations with a neural network representation.<n>We propose a more sample-efficient data-augmentation strategy for generating neural PDE training data from a computer model.
  arXiv  Detail & Related papers  (2025-08-26T21:22:11Z)
• Reasoning to Learn from Latent Thoughts [61.2395150828168]
  We show that explicitly modeling and inferring the emphlatent thoughts that underlie the text generation process can significantly improve pretraining data efficiency.<n>We show that a 1B LM can bootstrap its performance across at least three iterations and significantly outperform baselines trained on raw data.
  arXiv  Detail & Related papers  (2025-03-24T16:41:23Z)
• Model-Agnostic Knowledge Guided Correction for Improved Neural Surrogate Rollout [3.006104092368596]
  We propose a model-agnostic, RL based, cost-aware model which combines a neural surrogate, RL decision model, and a physics simulator to reduce surrogate rollout error significantly.<n>HyPER learns an intelligent policy that is adaptable to changing physical conditions and resistant to noise corruption.
  arXiv  Detail & Related papers  (2025-03-13T05:00:23Z)
• Improved Long Short-Term Memory-based Wastewater Treatment Simulators for Deep Reinforcement Learning [0.0]
  We implement two methods to improve the trained models for wastewater treatment data. The experimental results showed that implementing these methods can improve the behavior of simulators in terms of Dynamic Time Warping throughout a year.
  arXiv  Detail & Related papers  (2024-03-22T10:20:09Z)
• Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning [45.78096783448304]
  In this work, seeking data efficiency, we design unsupervised pretraining for PDE operator learning.<n>We mine unlabeled PDE data without simulated solutions, and we pretrain neural operators with physics-inspired reconstruction-based proxy tasks.<n>Our method is highly data-efficient, more generalizable, and even outperforms conventional vision-pretrained models.
  arXiv  Detail & Related papers  (2024-02-24T06:27:33Z)
• Training Deep Surrogate Models with Large Scale Online Learning [48.7576911714538]
  Deep learning algorithms have emerged as a viable alternative for obtaining fast solutions for PDEs. Models are usually trained on synthetic data generated by solvers, stored on disk and read back for training. It proposes an open source online training framework for deep surrogate models.
  arXiv  Detail & Related papers  (2023-06-28T12:02:27Z)
• Monte Carlo Neural PDE Solver for Learning PDEs via Probabilistic Representation [59.45669299295436]
  We propose a Monte Carlo PDE solver for training unsupervised neural solvers.<n>We use the PDEs' probabilistic representation, which regards macroscopic phenomena as ensembles of random particles.<n>Our experiments on convection-diffusion, Allen-Cahn, and Navier-Stokes equations demonstrate significant improvements in accuracy and efficiency.
  arXiv  Detail & Related papers  (2023-02-10T08:05:19Z)
• Continual learning autoencoder training for a particle-in-cell simulation via streaming [52.77024349608834]
  upcoming exascale era will provide a new generation of physics simulations with high resolution. These simulations will have a high resolution, which will impact the training of machine learning models since storing a high amount of simulation data on disk is nearly impossible. This work presents an approach that trains a neural network concurrently to a running simulation without data on a disk.
  arXiv  Detail & Related papers  (2022-11-09T09:55:14Z)
• Large-scale Neural Solvers for Partial Differential Equations [48.7576911714538]
  Solving partial differential equations (PDE) is an indispensable part of many branches of science as many processes can be modelled in terms of PDEs. Recent numerical solvers require manual discretization of the underlying equation as well as sophisticated, tailored code for distributed computing. We examine the applicability of continuous, mesh-free neural solvers for partial differential equations, physics-informed neural networks (PINNs) We discuss the accuracy of GatedPINN with respect to analytical solutions -- as well as state-of-the-art numerical solvers, such as spectral solvers.
  arXiv  Detail & Related papers  (2020-09-08T13:26:51Z)
• Designing Accurate Emulators for Scientific Processes using Calibration-Driven Deep Models [33.935755695805724]
  Learn-by-Calibrating (LbC) is a novel deep learning approach for designing emulators in scientific applications. We show that LbC provides significant improvements in generalization error over widely-adopted loss function choices. LbC achieves high-quality emulators even in small data regimes and more importantly, recovers the inherent noise structure without any explicit priors.
  arXiv  Detail & Related papers  (2020-05-05T16:54:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.