cs.LGNov 7, 2026

Spectral Bias Reduction in Local Gated ReLU Architectures

Authors: Victorita Dolean

Organizations: Department of Mathematics and Computer Science, Eindhoven University of Technology, PO Box 513, Eindhoven, 5600 MB, The Netherlands

Abstract

Neural networks trained by gradient-based methods learn the low-frequency content of a target function faster than its high-frequency content. This spectral bias is well documented, but it is much less clear which architectural choices mitigate it and which cannot. We study the question for shallow ReLU models linear in their trainable parameters, where each mode is learned at a rate given by a Gram-matrix eigenvalue, and for a general symmetric bilinear form a(⋅,⋅)a(\cdot,\cdot), so that least-squares regression (a=a= the L2L^2 inner product) and variational or physics-informed losses (a=a= the H1H^1 energy form) are two instances of one analysis. Within this framework we propose gated local ReLU dictionaries, obtained by multiplying each ReLU feature by a hat partition-of-unity weight, in the spirit of partition-of-unity finite elements and overlapping Schwarz methods. The resulting Gram matrices are block-banded for any local bilinear form, with a block-Toeplitz interior whose matrix-valued symbol governs the spectrum. The two forms then behave very differently. For L2L^2 regression, the symbol is bounded away from zero, the ratio is uniform in the number of windows, and the measured spread of frequency-dependent learning rates falls from 1.9×1051.9\times10^{5} to 4.54.5 at matched model size. For the H1H^1 stiffness matrix, the symbol vanishes at zero frequency: gating leaves the Θ(N2)Θ(N^2) conditioning intact and inverts the frequency ordering instead of removing the bias. A preconditioner built from the same interior stencil will fix the spectral bias, showing that architecture alone is not enough. Experiments on 11D regression and Poisson problems confirm the dichotomy, and the regression effect persists when the hidden-layer parameters are trained as well.

Figures & tables

Explore similar work

Feb 27, 2024

Implicit Bias via Spectral Neural Networks and Non-linear Matrix Sensing

The phenomenon of implicit bias has attracted interest in recent years as a fundamental aspect of the remarkable generalizing ability of neural networks. In a nutshell, it entails that gradient descent dynamics in many neural nets, even without any explicit regularizer in the loss function, converges to the solution of a regularized learning problem. In this work, we show that this phenomenon extends to non-linear spectral parameterizations in matrix sensing. More precisely, we propose a matrix network architecture called Spectral Neural Networks (abbrv. SNN) where each neuron applies nonlinear activation to the singular values of the input matrix. Conceptually, this entails coordinatizing the space of matrices by their singular values and singular vectors, as opposed to by their entries, a potentially fruitful perspective for matrix learning. We prove the exponentially fast decay of squared loss during the training dynamics of an SNN parameterized matrix sensing model under gradient flow (which is a continuous approximation of the discrete gradient descent), and also that the training implicitly selects an approximate solution that minimizes nuclear norm. We also demonstrate the effectiveness of the SNN architecture in the context of matrix sensing by empirical investigations. We believe that the SNN architecture has the potential to be of wide applicability in a broad class of matrix learning scenarios.
Mar 5, 2026

How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?

Overparameterized ML models, including neural networks, typically induce underdetermined training objectives with multiple global minima. The implicit bias refers to the limiting global minimum that is attained by a common optimization algorithm, such as gradient descent (GD). In this paper, we characterize the implicit bias of GD for training a shallow ReLU model with the squared loss on high-dimensional random features. Prior work (Vardi and Shamir, 2021) showed that the implicit bias does not exist in the worst-case, or corresponds exactly to the minimum-ℓ2\ell_2-norm interpolating solution under exactly orthogonal data (Boursier et al., 2022). Our work interpolates between these two extremes and shows that, for sufficiently high-dimensional random data, the implicit bias approximates the minimum-ℓ2\ell_2-norm solution with high probability with a gap on the order Θ(n/∣∣λ∣∣1)Θ(\sqrt{n/||λ||_1}), where nn is the number of training examples and λλ denotes the spectrum of the data covariance matrix. Our results are obtained through a novel primal-dual analysis that carefully tracks the evolution of predictions, data-span coefficients, as well as their interactions, and show that the ReLU activation pattern quickly stabilizes with high probability over random data.
Feb 22, 2026

Spectral bias in physics-informed and operator learning: Analysis and mitigation guidelines

Solving partial differential equations (PDEs) by neural networks as well as Kolmogorov-Arnold Networks (KANs), including physics-informed neural networks (PINNs), physics-informed KANs (PIKANs), and neural operators, are known to exhibit spectral bias, whereby low-frequency components of the solution are learned significantly faster than high-frequency modes. While spectral bias is often treated as an intrinsic representational limitation of neural architectures, its interaction with optimization dynamics and physics-based loss formulations remains poorly understood. In this work, we provide a systematic investigation of spectral bias in physics-informed and operator learning frameworks, with emphasis on the coupled roles of network architecture, activation functions, loss design, and optimization strategy. We quantify spectral bias through frequency-resolved error metrics, Barron-norm diagnostics, and higher-order statistical moments, enabling a unified analysis across elliptic, hyperbolic, and dispersive PDEs. Through diverse benchmark problems, including the Korteweg-de Vries, wave and steady-state diffusion-reaction equations, turbulent flow reconstruction, and earthquake dynamics, we demonstrate that spectral bias is not simply representational but fundamentally dynamical. In particular, second-order optimization methods substantially alter the spectral learning order, enabling earlier and more accurate recovery of high-frequency modes for all PDE types. For neural operators, we further show that spectral bias is dependent on the neural operator architecture and can also be effectively mitigated through spectral-aware loss formulations without increasing the inference cost.