2018
DOI: 10.1109/tnnls.2017.2650865
|View full text |Cite
|
Sign up to set email alerts
|

Multicolumn RBF Network

Abstract: This paper proposes the multicolumn RBF network (MCRN) as a method to improve the accuracy and speed of a traditional radial basis function network (RBFN). The RBFN, as a fully connected artificial neural network (ANN), suffers from costly kernel inner-product calculations due to the use of many instances as the centers of hidden units. This issue is not critical for small datasets, as adding more hidden units will not burden the computation time. However, for larger datasets, the RBFN requires many hidden uni… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
8
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
9

Relationship

0
9

Authors

Journals

citations
Cited by 28 publications
(8 citation statements)
references
References 28 publications
0
8
0
Order By: Relevance
“…To design an appropriate estimation and compensation algorithm, Gaussian radial basis function neural networks (RBFNNs) are an alternative method; by virtue of the simple topological structure and universal approximation ability, RBFNNs have been widely applied for nonlinear system modeling and control [38][39][40][41][42][43][44][45][46]. In this paper, RBFNNs are used to estimate and compensate for all the nonperiodic uncertainties of the robot manipulator system.…”
Section: Preliminariesmentioning
confidence: 99%
“…To design an appropriate estimation and compensation algorithm, Gaussian radial basis function neural networks (RBFNNs) are an alternative method; by virtue of the simple topological structure and universal approximation ability, RBFNNs have been widely applied for nonlinear system modeling and control [38][39][40][41][42][43][44][45][46]. In this paper, RBFNNs are used to estimate and compensate for all the nonperiodic uncertainties of the robot manipulator system.…”
Section: Preliminariesmentioning
confidence: 99%
“…Moreover, some efficient and effective training approaches for RBFN have been proposed recently, which may also be extended to TSK fuzzy system. For example, multicolumn RBFN [32], which divides a large dataset into smaller subsets using the k-d tree algorithm and then trains an RBFN for each subset, has demonstrated faster speed and higher accuracy than the traditional RBFN. This approach could be extended to TSK fuzzy systems for big data problems.…”
Section: B Functional Equivalence Between Tsk Fuzzy Systems and Radia...mentioning
confidence: 99%
“…Recently, lightweight learning models have also been proposed for hyperspectral image classification such as lightweight convolutional neural network [22] and spectralspatial squeeze-and-excitation residual bag-of-features learning [23]. Another way to reduce the training time is using RBF network [24,25,26,27]. It has two layers including the kernal fuctions layer and the fully connected layer.…”
Section: Introductionmentioning
confidence: 99%
“…The recent works on RBF networks can be categorized into two types. (1) Using RBF networks as an ensemble: For example, multicolumn RBF network [25] uses k-d tree two select subsets of features, and learns these subsets with different RBF networks. It has excellent performance on classification of manually selected features, but it is difficult to use this kind of ensemble to do image classification.…”
Section: Introductionmentioning
confidence: 99%