DocumentCode :
303814
Title :
Merging information in the data and weight spaces
Author :
Burrascano, Pietro ; Pirollo, Dario
Author_Institution :
Inst. of Electron., Perugia Univ., Italy
Volume :
2
fYear :
1996
fDate :
13-16 May 1996
Firstpage :
617
Abstract :
The paper addresses the problem of combining independent information which can be available in both the data and parameters spaces: the objective is to obtain a neural model which takes into account the information available from both sources. The problem is approached in the framework of the probabilistic interpretation of neural modelling and the indetermination associated to the training process is taken into account by considering an appropriate distribution in the weight space associated to each solution vector. A computationally light procedure is proposed to merge the information associated to the different solutions. The effectiveness of the proposed procedure is shown by means of experiments of feedforward neural networks for classification tasks
Keywords :
learning (artificial intelligence); neural nets; pattern classification; classification tasks; data space; feedforward neural network; neural model; neural modelling; probabilistic interpretation; weight space; Cost function; Covariance matrix; Equations; Extraterrestrial measurements; Gaussian distribution; Intelligent networks; Merging; Optimized production technology; Parametric statistics; Volume measurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrotechnical Conference, 1996. MELECON '96., 8th Mediterranean
Conference_Location :
Bari
Print_ISBN :
0-7803-3109-5
Type :
conf
DOI :
10.1109/MELCON.1996.551296
Filename :
551296
Link To Document :
بازگشت