| Convergence and objective functions of some fault/noise-injection-based online learning algorithms for RBF networks. | |
| | |
MedLine Citation:
|
PMID: 20388593 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
In the last two decades, many online fault/noise injection algorithms have been developed to attain a fault tolerant neural network. However, not much theoretical works related to their convergence and objective functions have been reported. This paper studies six common fault/noise-injection-based online learning algorithms for radial basis function (RBF) networks, namely 1) injecting additive input noise, 2) injecting additive/multiplicative weight noise, 3) injecting multiplicative node noise, 4) injecting multiweight fault (random disconnection of weights), 5) injecting multinode fault during training, and 6) weight decay with injecting multinode fault. Based on the Gladyshev theorem, we show that the convergence of these six online algorithms is almost sure. Moreover, their true objective functions being minimized are derived. For injecting additive input noise during training, the objective function is identical to that of the Tikhonov regularizer approach. For injecting additive/multiplicative weight noise during training, the objective function is the simple mean square training error. Thus, injecting additive/multiplicative weight noise during training cannot improve the fault tolerance of an RBF network. Similar to injective additive input noise, the objective functions of other fault/noise-injection-based online algorithms contain a mean square error term and a specialized regularization term. |
| | |
Authors:
|
Kevin I-J Ho; Chi-Sing Leung; John Sum |
Related Documents
:
|
20330413 - Fan noise prediction for design. 17155043 - Noise accelerates synchronization of coupled nonlinear oscillators. 22048383 - Optical manipulation of microparticles using whispering-gallery modes in a silicon nitr... |
Publication Detail:
|
Type: Journal Article; Research Support, Non-U.S. Gov't Date: 2010-04-12 |
Journal Detail:
|
Title: IEEE transactions on neural networks / a publication of the IEEE Neural Networks Council Volume: 21 ISSN: 1941-0093 ISO Abbreviation: IEEE Trans Neural Netw Publication Date: 2010 Jun |
Date Detail:
|
Created Date: 2010-07-27 Completed Date: 2010-11-12 Revised Date: - |
Medline Journal Info:
|
Nlm Unique ID: 101211035 Medline TA: IEEE Trans Neural Netw Country: United States |
Other Details:
|
Languages: eng Pagination: 938-47 Citation Subset: IM |
Affiliation:
|
Department of Computer Science and Communication Engineering, Providence University, Sha-Lu 433, Taiwan. ho@pu.edu.tw |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Algorithms* Humans Learning* Neural Networks (Computer)* Online Systems* Signal Processing, Computer-Assisted* |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Effect of MR distortion on targeting for deep-brain stimulation.
Next Document: A New One-Layer Neural Network for Linear and Quadratic Programming.