A novel kernel for RBF based neural networks
Summary: Radial basis function (RBF) is well known to provide excellent performance in function approximation and pattern classification. The conventional RBF uses basis functions which rely on distance measures such as Gaussian kernel of Euclidean distance (ED) between feature vector and neuron's center, and so forth. In this work, we introduce a novel RBF artificial neural network (ANN) where the basis function utilizes a linear combination of ED based Gaussian kernel and a cosine kernel where the cosine kernel computes the angle between feature and center vectors. Novelty of the proposed work relies on the fact that we have shown that there may be scenarios where the two feature vectors (FV) are more prominently distinguishable via the proposed cosine measure as compared to the conventional ED measure. We discuss adaptive symbol detection for multiple phase shift keying (MPSK) signals as a practical example to show where the angle information can be pivotal which in turn justifies our proposed RBF kernel. To corroborate our theoretical developments, we investigate the performance of the proposed RBF for the problems pertaining to three different domains. Our results show that the proposed RBF outperforms the conventional RBF by a remarkable margin.
- scientific article; zbMATH DE number 1960356
- scientific article; zbMATH DE number 2079453
- Kernel Based Learning Methods: Regularization Networks and RBF Networks
- Computational and Information Science
- A Novel Continuous Forward Algorithm for RBF Neural Modelling
- A multi-scale radial basis function network with multiple kernels
- Nonlinear function learning, and classification using RBF networks with optimal kernels
- A logical calculus of the ideas immanent in nervous activity
- Feature selection for fault level diagnosis of planetary gearboxes
- Geometrical and Statistical Properties of Systems of Linear Inequalities with Applications in Pattern Recognition
- scientific article; zbMATH DE number 4074523 (Why is no real title available?)
- scientific article; zbMATH DE number 51537 (Why is no real title available?)
- Learning representations by back-propagating errors
- Neocognition: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position
- Silhouettes: a graphical aid to the interpretation and validation of cluster analysis
- Gaussian kernel approximation algorithm for feedforward neural network design
- Diversity combining for digital mobile radio using radial basis function networks
- Predictive modeling based on small data in clinical medicine: RBF-based additive input-doubling method
- scientific article; zbMATH DE number 1960356 (Why is no real title available?)
- scientific article; zbMATH DE number 2079453 (Why is no real title available?)
- scientific article; zbMATH DE number 7426461 (Why is no real title available?)
- A new variable shape parameter strategy for RBF approximation using neural networks
This page was built for publication: A novel kernel for RBF based neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1722266)
