When Does Frequency Decomposition Benefit Physics-Informed Neural Networks? A Preliminary Ablation Study

• URL: http://arxiv.org/abs/2608.24940v1
• Date: Mon, 24 Aug 2026 09:48:16 GMT
• Title: When Does Frequency Decomposition Benefit Physics-Informed Neural Networks? A Preliminary Ablation Study
• Abstract summary: Partial differential equations (PDEs) often have high-frequency and multi-scale features that neural networks struggle to approximate.<n>PINNs build the governing equations directly into training, but suffer from spectral bias.<n>We introduce a dual-branch, spectrally-gated architecture (DBSG-PINN) that splits low- and high-frequency components into separateworks joined by an adaptive gate.
• Score: 1.2081877372552605
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Partial differential equations (PDEs) often have high-frequency and multi-scale features that neural networks struggle to approximate. Physics-Informed Neural Networks (PINNs) build the governing equations directly into training, but suffer from spectral bias: they learn low-frequency components faster than high-frequency ones. Techniques such as Fourier feature embeddings and sinusoidal activations address this, but most studies assume they help across the board without checking which spectral regimes actually benefit. We introduce a dual-branch, spectrally-gated architecture (DBSG-PINN) that splits low- and high-frequency components into separate subnetworks joined by an adaptive gate, and use it to run a partially controlled ablation of frequency decomposition and spectral routing. We test this on five one-dimensional benchmark PDEs, ranging from smooth, single-scale problems to oscillatory, multi-scale ones. Frequency decomposition helps most on the spectrally complex benchmarks, cutting relative $L_2$ error by up to $59.2\%$ on a multimodal wave problem, but gives little benefit on smoother PDEs. On one benchmark (1D Wave), it performs substantially worse than a simpler fixed-combination variant. The gate's benefit scales with how spectrally rich the target solution is: the full model's advantage over the ablations is largest on multi-scale benchmarks and smallest (or negative) on single-scale ones, consistent with the gate exploiting frequency structure rather than acting as noise,though we do not directly visualize or quantify its spatial activations in this study. All results come from a single training seed across five 1D benchmarks, so we present this as an exploratory study meant to raise questions rather than answer them, and outline the additional seeds and benchmarks needed to test whether the pattern holds.

Related papers

• Frequency Shift Physics-Informed Extreme Learning Machine for Solving High-Frequency Partial Differential Equations [6.85637056992485]
  Solving partial differential equations with high-frequency solutions remains a central challenge in physics-informed machine learning.<n>This paper proposes a Frequency Shift Physics-Informed Extreme Learning Machine framework that addresses this limitation.<n>Two variants are developed: FS-PIELM-L assigns independent frequency magnitudes to individual neurons, while FS-PIELM-G groups neurons for improved robustness.
  arXiv  Detail & Related papers  (2026-07-02T04:37:40Z)
• Closed-Form Spectral Regularization for Multi-Task Model Merging [96.82449201305234]
  Model merging combines several independently fine-tuned experts into a single multi-task model without any training data.<n>State-of-the-art merging methods formulate merging as a layer-wise interference problem.<n>We propose SWUDI, a closed-form method that combines a soft exponential filter, which matches the gradient-flow trajectory of iterative descent.
  arXiv  Detail & Related papers  (2026-06-05T14:00:47Z)
• The Adaptive Vekua Cascade: A Differentiable Spectral-Analytic Solver for Physics-Informed Representation [0.0]
  Adaptive-based neural networks have emerged as a powerful tool for representing continuous physical fields.<n>They face two fundamental pathologies: spectral bias and the curse of dimensionality.<n>We propose a hybrid architecture that bridges analytic analytic decouples manifold learning from function approximation.
  arXiv  Detail & Related papers  (2025-12-12T18:41:35Z)
• LSCD: Lomb-Scargle Conditioned Diffusion for Time series Imputation [55.800319453296886]
  Time series with missing or irregularly sampled data are a persistent challenge in machine learning.<n>We introduce a different Lombiable--Scargle layer that enables a reliable computation of the power spectrum of irregularly sampled data.
  arXiv  Detail & Related papers  (2025-06-20T14:48:42Z)
• LOGLO-FNO: Efficient Learning of Local and Global Features in Fourier Neural Operators [20.77877474840923]
  High-frequency information is a critical challenge in machine learning.<n>Deep neural nets exhibit the so-called spectral bias toward learning low-frequency components.<n>We propose a novel frequency-sensitive loss term based on radially binned spectral errors.
  arXiv  Detail & Related papers  (2025-04-05T19:35:04Z)
• Constructing Extreme Learning Machines with zero Spectral Bias [0.0]
  Extreme Learning Machines (ELMs) obviate an iterative solution process which provides the theoretical basis of Spectral Bias (SB) This work verifies the reliability of this assumption, and shows that it is incorrect. The structure of ELMs makes them naturally amenable to implementation of variants of Fourier Feature Embeddings, which have been shown to mitigate SB in ANNs. This approach is implemented and verified to completely eliminate SB, thus bringing into feasibility the application of ELMs for practical problems like PINNs where resolution of higher frequencies is essential.
  arXiv  Detail & Related papers  (2023-07-19T05:41:40Z)
• One-Dimensional Deep Image Prior for Curve Fitting of S-Parameters from Electromagnetic Solvers [57.441926088870325]
  Deep Image Prior (DIP) is a technique that optimized the weights of a randomly-d convolutional neural network to fit a signal from noisy or under-determined measurements. Relative to publicly available implementations of Vector Fitting (VF), our method shows superior performance on nearly all test examples.
  arXiv  Detail & Related papers  (2023-06-06T20:28:37Z)
• Incremental Spatial and Spectral Learning of Neural Operators for Solving Large-Scale PDEs [86.35471039808023]
  We introduce the Incremental Fourier Neural Operator (iFNO), which progressively increases the number of frequency modes used by the model. We show that iFNO reduces total training time while maintaining or improving generalization performance across various datasets. Our method demonstrates a 10% lower testing error, using 20% fewer frequency modes compared to the existing Fourier Neural Operator, while also achieving a 30% faster training.
  arXiv  Detail & Related papers  (2022-11-28T09:57:15Z)
• Transform Once: Efficient Operator Learning in Frequency Domain [69.74509540521397]
  We study deep neural networks designed to harness the structure in frequency domain for efficient learning of long-range correlations in space or time. This work introduces a blueprint for frequency domain learning through a single transform: transform once (T1)
  arXiv  Detail & Related papers  (2022-11-26T01:56:05Z)
• Momentum Diminishes the Effect of Spectral Bias in Physics-Informed Neural Networks [72.09574528342732]
  Physics-informed neural network (PINN) algorithms have shown promising results in solving a wide range of problems involving partial differential equations (PDEs) They often fail to converge to desirable solutions when the target function contains high-frequency features, due to a phenomenon known as spectral bias. In the present work, we exploit neural tangent kernels (NTKs) to investigate the training dynamics of PINNs evolving under gradient descent with momentum (SGDM)
  arXiv  Detail & Related papers  (2022-06-29T19:03:10Z)

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.