Kolmogorov-Arnold Networks are Radial Basis Function Networks
arXiv:2405.06721v1194 citationsh-index: 1Has Code
Originality Synthesis-oriented
AI Analysis
This is an incremental improvement for researchers and practitioners using KANs, as it addresses a computational bottleneck without altering the core paradigm.
The paper tackled the computational inefficiency of Kolmogorov-Arnold Networks (KANs) by approximating their 3-order B-splines with Gaussian radial basis functions, resulting in FastKAN, a faster implementation that is also a radial basis function network.
This short paper is a fast proof-of-concept that the 3-order B-splines used in Kolmogorov-Arnold Networks (KANs) can be well approximated by Gaussian radial basis functions. Doing so leads to FastKAN, a much faster implementation of KAN which is also a radial basis function (RBF) network.