Universal approximation capability of EBF neural networks with arbitrary activation functions

DOI10.1007/BF01188988zbMath0860.68086MaRDI QIDQ2563557

Hong Chen, Tian-Ping Chen

Publication date: 16 December 1996

Published in: Circuits, Systems, and Signal Processing (Search for Journal in Brave)

zbMATH Keywords

neural networksapproximation capabilityelliptic basis function

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items

Indirect adaptive control of nonlinear dynamic systems using self recurrent wavelet neural networks via adaptive learning rates ⋮ \(L^p\) approximation capability of RBF neural networks ⋮ The universal approximation capabilities of double \(2\pi\)-periodic approximate identity neural networks ⋮ Capability of neural networks in computing the outputs of dynamic systems with inputs defined on the whole space

Cites Work

• Unnamed Item
• Unnamed Item
• Structure theorems for nonlinear systems
• Approximation by superposition of sigmoidal and radial basis functions
• Nonlinear approximations using elliptic basis function networks
• Multilayer feedforward networks are universal approximators
• Implications and applications of Kolmogorov's superposition theorem
• Approximation by superpositions of a sigmoidal function