paper-with-me

홈 › Papers

Generalization Bounds and Model Complexity for Kolmogorov-Arnold Networks

2024-10-10 · Xianyang Zhang, Huijuan Zhou

Kolmogorov-Arnold Network (KAN) is a network structure recently proposed by Liu et al. (2024) that offers improved interpretability and a more parsimonious design in many science-oriented tasks compared to multi-layer perceptrons. This work provides a rigorous theoretical analysis of KAN by establishing generalization bounds for KAN equipped with activation functions that are either represented by linear combinations of basis functions or lying in a low-rank Reproducing Kernel Hilbert Space (RKHS). In the first case, the generalization bound accommodates various choices of basis functions in forming the activation functions in each layer of KAN and is adapted to different operator norms at each layer. For a particular choice of operator norms, the bound scales with the $l_1$ norm of the coefficient matrices and the Lipschitz constants for the activation functions, and it has no dependence on combinatorial parameters (e.g., number of nodes) outside of logarithmic factors. Moreover, our result does not require the boundedness assumption on the loss function and, hence, is applicable to a general class of regression-type loss functions. In the low-rank case, the generalization bound scales polynomially with the underlying ranks as well as the Lipschitz constants of the activation functions in each layer. These bounds are empirically investigated for KANs trained with stochastic gradient descent on simulated and real data sets. The numerical results demonstrate the practical relevance of these bounds.

📄 PDF Abstract BibTeX arXiv:2410.08026

Code (0)

등록된 구현이 없습니다.

Tasks

Generalization BoundsKolmogorov-Arnold Networks

Methods 이 논문이 사용한 방법론

+ ( 1 ) ⟷ 805 ⟷ ( 330 ) ⟷ 4056|How do I file a complaint with Expedia? How do I file a complaint with Expedia contact customer service at + ( 1 ) ⟷ 888 ⟷ ( 829 ) ⟷ 0881 or + ( 1 ) ⟷ 805 ⟷ ( 330 ) ⟷ 4056, or use their Help Center. Explain your issue…

Similar Papers 제목 키워드 기반

KKANs: Kurkova-Kolmogorov-Arnold Networks and Their Learning Dynamics

2024-12-21 · Juan Diego Toscano, Li-Lian Wang, George Em Karniadakis

Inspired by the Kolmogorov-Arnold representation theorem and Kurkova's principle of using approximate representations, we propose the Kurkova-Kolmogorov-Arnold Network (KKAN), a new two-block architecture that combines r…

Kolmogorov-Arnold NetworksOperator learningPhysics-informed machine learning

Error bounds for deep ReLU networks using the Kolmogorov--Arnold superposition theorem

2019-06-27 · Hadrien Montanelli, Haizhao Yang

We prove a theorem concerning the approximation of multivariate functions by deep ReLU networks, for which the curse of the dimensionality is lessened. Our theorem is based on a constructive proof of the Kolmogorov--Arno…

P1-KAN: an effective Kolmogorov-Arnold network with application to hydraulic valley optimization

2024-10-04 · Xavier Warin

A new Kolmogorov-Arnold network (KAN) is proposed to approximate potentially irregular functions in high dimensions. We provide error bounds for this approximation, assuming that the Kolmogorov-Arnold expansion functions…

Optimization, Generalization and Differential Privacy Bounds for Gradient Descent on Kolmogorov-Arnold Networks

2026-01-29 · Puyu Wang, Junyu Zhou, Philipp Liznerski, Marius Kloft arxiv

Kolmogorov--Arnold Networks (KANs) have recently emerged as a structured alternative to standard MLPs, yet a principled theory for their training dynamics, generalization, and privacy properties remains limited. In this …

Non-linear Equalization in 112 Gb/s PONs Using Kolmogorov-Arnold Networks

2024-11-29 · Rodrigo Fischer, Patrick Matalla, Sebastian Randel, Laurent Schmalen

We investigate Kolmogorov-Arnold networks (KANs) for non-linear equalization of 112 Gb/s PAM4 passive optical networks (PONs). Using pruning and extensive hyperparameter search, we outperform linear equalizers and convol…

Kolmogorov-Arnold Networks