aiqbench analysis
This research introduces PH-KAN, a novel framework that integrates Kolmogorov-Arnold Networks into port-Hamiltonian system identification to preserve physical structures while enhancing model interpretability. By replacing standard multilayer perceptrons with dedicated KAN blocks for specific constitutive components, the method offers explicit inspection of nonlinear functions in physics-informed models. This advancement addresses a key limitation in data-driven scientific machine learning where traditional black-box architectures fail to maintain underlying dynamical constraints.
PH-KAN: Port-Hamiltonian Kolmogorov-Arnold Network
arXiv:2606.14708v1 Announce Type: cross Abstract: Data-driven machine learning approaches have become increasingly attractive for nonlinear system identification, but standard models often fail to preserve the underlying physical structure and remain difficult to interpret, especially when no analytical model is available. In this context, port-Hamiltonian (pH) models provide a natural physics-informed representation. However, when these models are parameterized with standard multilayer perceptrons (MLPs), the learned constitutive components often remain poorly interpretable. In this paper, we propose a structure-preserving identification framework for nonlinear port-Hamiltonian systems based on Kolmogorov-Arnold Networks (KANs). The proposed PH-KAN model parameterizes the interconnection matrix, dissipation matrix, Hamiltonian, and input mapping using dedicated KAN blocks, while enforcing the port-Hamiltonian constraints by construction. This yields constitutive representations in which the nonlinear functions defining the identified pH components can be explicitly inspected, leading to a more interpretable model than with standard MLP-based parameterizations.
Abstract:Data-driven machine learning approaches have become increasingly attractive for nonlinear system identification, but standard models often fail to preserve the underlying physical structure and remain difficult to interpret, especially when no analytical model is available. In this context, port-Hamiltonian (pH) models provide a natural physics-informed representation. However, when these models are parameterized with standard multilayer perceptrons (MLPs), the learned constitutive components often remain poorly interpretable. In this paper, we propose a structure-preserving identification framework for nonlinear port-Hamiltonian systems based on Kolmogorov-Arnold Networks (KANs). The proposed PH-KAN model parameterizes the interconnection matrix, dissipation matrix, Hamiltonian, and input mapping using dedicated KAN blocks, while enforcing the port-Hamiltonian constraints by construction. This yields constitutive representations in which the nonlinear functions defining the identified pH components can be explicitly inspected, leading to a more interpretable model than with standard MLP-based parameterizations.
Optimization and Control (math.OC); Artificial Intelligence (cs.AI)
arXiv:2606.14708 [math.OC]
(or arXiv:2606.14708v1 [math.OC] for this version)
Full text under the publisher's open licence · 152 words