English   español  
Please use this identifier to cite or link to this item: http://hdl.handle.net/10261/30523
Share/Impact:
Title: A deterministic algorithm that emulates learning with random weights
Authors: Ruiz de Angulo, Vicente; Torras, Carme
Keywords: Noise
Noisy training
Regularization
Fault-tolerance
Optimization
Sensitivity
Adaptive systems
Automation
Issue Date: 2002
Publisher: Elsevier
Citation: Neurocomputing 48(1): 975-1002 (2002)
Abstract: The expectation of a function of random variables can be modeled as the value of the function in the mean value of the variables plus a penalty term. Here, this penalty term is calculated exactly, and the properties of different approximations are analyzed. Then, a deterministic algorithm for minimizing the expected error of a feedforward network of random weights is presented. Given a particular feedforward network architecture and a training set, this algorithm accurately finds the weight configuration that makes the network response most resistant to a class of weight perturbations. Finally, the study of the most stable configurations of a network unravels some undesirable properties of networks with asymmetric activation functions.
Publisher version (URL): http://dx.doi.org/10.1016/S0925-2312(01)00695-6
URI: http://hdl.handle.net/10261/30523
DOI: 10.1016/S0925-2312(01)00695-6
ISSN: 0925-2312
Appears in Collections:(IRII) Artículos
Files in This Item:
File Description SizeFormat 
doc1.pdf287,98 kBAdobe PDFThumbnail
View/Open
Show full item record
 



WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.