Towards Deep Physics-Informed Kolmogorov-Arnold Networks

• URL: http://arxiv.org/abs/2510.23501v1
• Date: Mon, 27 Oct 2025 16:35:01 GMT
• Title: Towards Deep Physics-Informed Kolmogorov-Arnold Networks
• Abstract summary: Chebyshev-based physics-informed KANs (cPIKANs) have become the standard due to their computational efficiency.<n>We propose a basis-agnostic, Glorot-like scheme that preserves activation variance and yields substantial improvements in stability and accuracy.<n>Inspired by the PirateNet architecture, we introduce Residual-Gated Adaptive KANs (RGA KANs)<n>We show that RGA KANs successfully traverse all training phases, unlike baseline cPIKANs, which stagnate in the diffusion phase in specific PDE settings.
• Score: 4.725935825821886
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Since their introduction, Kolmogorov-Arnold Networks (KANs) have been successfully applied across several domains, with physics-informed machine learning (PIML) emerging as one of the areas where they have thrived. In the PIML setting, Chebyshev-based physics-informed KANs (cPIKANs) have become the standard due to their computational efficiency. However, like their multilayer perceptron-based counterparts, cPIKANs face significant challenges when scaled to depth, leading to training instabilities that limit their applicability to several PDE problems. To address this, we propose a basis-agnostic, Glorot-like initialization scheme that preserves activation variance and yields substantial improvements in stability and accuracy over the default initialization of cPIKANs. Inspired by the PirateNet architecture, we further introduce Residual-Gated Adaptive KANs (RGA KANs), designed to mitigate divergence in deep cPIKANs where initialization alone is not sufficient. Through empirical tests and information bottleneck analysis, we show that RGA KANs successfully traverse all training phases, unlike baseline cPIKANs, which stagnate in the diffusion phase in specific PDE settings. Evaluations on seven standard forward PDE benchmarks under a fixed training pipeline with adaptive components demonstrate that RGA KANs consistently outperform parameter-matched cPIKANs and PirateNets - often by several orders of magnitude - while remaining stable in settings where the others diverge.

Related papers

• LIGO-PINN: Learned Initialization via Gated Optimization to Alleviate Convergence Failures in Physics Informed Neural Networks [4.769951565159109]
  Physics-informed neural networks (PINNs) have had a broad research impact in modeling domains governed by partial differential equations (PDE)<n> PINNs have been shown to perform poorly, sometimes even converging to trivial solutions, in challenging PDE domains.<n>We propose a framework for Learned Initialization via Gated Layerwise Optimization (LIGO-PINN) to overcome PINN convergence failures.
  arXiv  Detail & Related papers  (2026-07-15T18:00:44Z)
• Kernel-Adaptive PI-ELMs for Forward and Inverse Problems in PDEs with Sharp Gradients [0.0]
  This paper introduces the Kernel Adaptive Physics-Informed Extreme Learning Machine (KAPI-ELM)<n>It is designed to solve both forward and inverse Partial Differential Equation (PDE) problems involving localized sharp gradients.<n>KAPI-ELM achieves state-of-the-art accuracy in both forward and inverse settings.
  arXiv  Detail & Related papers  (2025-07-14T13:03:53Z)
• AC-PKAN: Attention-Enhanced and Chebyshev Polynomial-Based Physics-Informed Kolmogorov-Arnold Networks [3.6190123930006317]
  We present AC-PKAN, a novel architecture that constitutes an enhancement to weakly supervised Physics-Informed Neural Networks (PINNs)<n>We show that AC-PKAN consistently outperforms or matches state-of-the-art models such as PINNsFormer.
  arXiv  Detail & Related papers  (2025-05-13T15:46:10Z)
• Mitigating Barren Plateaus in Quantum Neural Networks via an AI-Driven Submartingale-Based Framework [3.0617189749929348]
  We propose AdaInit to mitigate barren plateaus (BPs) in quantum neural networks (QNNs)<n>AdaInit iteratively synthesizes initial parameters for QNNs that yield non-negligible gradient variance, thereby mitigating BPs.<n>We provide rigorous theoretical analyses of the submartingale-based process and empirically validate that AdaInit consistently outperforms existing methods in maintaining higher gradient variance across various QNN scales.
  arXiv  Detail & Related papers  (2025-02-17T05:57:15Z)
• Advancing Generalization in PINNs through Latent-Space Representations [71.86401914779019]
  Physics-informed neural networks (PINNs) have made significant strides in modeling dynamical systems governed by partial differential equations (PDEs)<n>We propose PIDO, a novel physics-informed neural PDE solver designed to generalize effectively across diverse PDE configurations.<n>We validate PIDO on a range of benchmarks, including 1D combined equations and 2D Navier-Stokes equations.
  arXiv  Detail & Related papers  (2024-11-28T13:16:20Z)
• Adaptive Training of Grid-Dependent Physics-Informed Kolmogorov-Arnold Networks [4.216184112447278]
  Physics-Informed Neural Networks (PINNs) have emerged as a robust framework for solving Partial Differential Equations (PDEs) We present a fast JAX-based implementation of grid-dependent Physics-Informed Kolmogorov-Arnold Networks (PIKANs) for solving PDEs. We demonstrate that the adaptive features significantly enhance solution accuracy, decreasing the L2 error relative to the reference solution by up to 43.02%.
  arXiv  Detail & Related papers  (2024-07-24T19:55:08Z)
• RoPINN: Region Optimized Physics-Informed Neural Networks [66.38369833561039]
  Physics-informed neural networks (PINNs) have been widely applied to solve partial differential equations (PDEs) This paper proposes and theoretically studies a new training paradigm as region optimization. A practical training algorithm, Region Optimized PINN (RoPINN), is seamlessly derived from this new paradigm.
  arXiv  Detail & Related papers  (2024-05-23T09:45:57Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)
• A Gaussian Process Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations [0.0]
  We introduce kernel-weighted Corrective Residuals (CoRes) to integrate the strengths of kernel methods and deep NNs for solving nonlinear PDE systems. CoRes consistently outperforms competing methods in solving a broad range of benchmark problems. We believe our findings have the potential to spark a renewed interest in leveraging kernel methods for solving PDEs.
  arXiv  Detail & Related papers  (2024-01-07T14:09:42Z)
• Deep Equilibrium Based Neural Operators for Steady-State PDEs [100.88355782126098]
  We study the benefits of weight-tied neural network architectures for steady-state PDEs. We propose FNO-DEQ, a deep equilibrium variant of the FNO architecture that directly solves for the solution of a steady-state PDE.
  arXiv  Detail & Related papers  (2023-11-30T22:34:57Z)
• PhyCRNet: Physics-informed Convolutional-Recurrent Network for Solving Spatiotemporal PDEs [8.220908558735884]
  Partial differential equations (PDEs) play a fundamental role in modeling and simulating problems across a wide range of disciplines. Recent advances in deep learning have shown the great potential of physics-informed neural networks (NNs) to solve PDEs as a basis for data-driven inverse analysis. We propose the novel physics-informed convolutional-recurrent learning architectures (PhyCRNet and PhCRyNet-s) for solving PDEs without any labeled data.
  arXiv  Detail & Related papers  (2021-06-26T22:22:19Z)
• Recent Developments Combining Ensemble Smoother and Deep Generative Networks for Facies History Matching [58.720142291102135]
  This research project focuses on the use of autoencoders networks to construct a continuous parameterization for facies models. We benchmark seven different formulations, including VAE, generative adversarial network (GAN), Wasserstein GAN, variational auto-encoding GAN, principal component analysis (PCA) with cycle GAN, PCA with transfer style network and VAE with style loss.
  arXiv  Detail & Related papers  (2020-05-08T21:32:42Z)
• Revisiting Initialization of Neural Networks [72.24615341588846]
  We propose a rigorous estimation of the global curvature of weights across layers by approximating and controlling the norm of their Hessian matrix. Our experiments on Word2Vec and the MNIST/CIFAR image classification tasks confirm that tracking the Hessian norm is a useful diagnostic tool.
  arXiv  Detail & Related papers  (2020-04-20T18:12:56Z)

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.