In the nonlinear systems, the preknowledge about the exact functional structure between inputs and outputs is mostly either unavailable or insufficient. In this case, the artificial neural networks (ANNs) are useful tools to estimate this functional structure. However, the traditional ANNs with the sum squared error suffer from the approximation and estimation errors in the high dimensional and excessive nonlinear cases. In this context, Bayesian neural networks (BNNs) provide a natural way to alleviate these issues by means of penalizing the excessive complex models. Thus, this approach allows estimating more reliable and robust models in the regression analysis, time series, pattern recognition problems etc. This paper presents a Bayesian learning approach based on Gaussian approximation which estimates the parameters and hyperparameters in the BNNs efficiently. In the application part, the proposed approach is compared with the traditional ANNs in terms of their estimation and prediction performances over an artificial data set.
Konular  Matematik ve İstatistik 

Yayımlanma Tarihi  Aralık 2017 
Dergi Bölümü  Articles 
Yazarlar 

Bibtex  @araştırma makalesi { forecasting346891,
journal = {Turkish Journal of Forecasting},
issn = {},
address = {Giresun University Forecast Research Laboratory},
year = {2017},
volume = {01},
pages = {54  65},
doi = {},
title = {Bayesian Learning based Gaussian Approximation for Artificial Neural Networks},
key = {cite},
author = {Koacadagli, Ozan}
} 
APA  Koacadagli, O . (2017). Bayesian Learning based Gaussian Approximation for Artificial Neural Networks. Turkish Journal of Forecasting, 01 (2), 5465. Retrieved from http://dergipark.gov.tr/forecasting/issue/33413/346891 
MLA  Koacadagli, O . "Bayesian Learning based Gaussian Approximation for Artificial Neural Networks". Turkish Journal of Forecasting 01 (2017): 5465 <http://dergipark.gov.tr/forecasting/issue/33413/346891> 
Chicago  Koacadagli, O . "Bayesian Learning based Gaussian Approximation for Artificial Neural Networks". Turkish Journal of Forecasting 01 (2017): 5465 
RIS  TY  JOUR T1  Bayesian Learning based Gaussian Approximation for Artificial Neural Networks AU  Ozan Koacadagli Y1  2017 PY  2017 N1  DO  T2  Turkish Journal of Forecasting JF  Journal JO  JOR SP  54 EP  65 VL  01 IS  2 SN  26186594 M3  UR  Y2  2017 ER  
EndNote  %0 Turkish Journal of Forecasting Bayesian Learning based Gaussian Approximation for Artificial Neural Networks %A Ozan Koacadagli %T Bayesian Learning based Gaussian Approximation for Artificial Neural Networks %D 2017 %J Turkish Journal of Forecasting %P 26186594 %V 01 %N 2 %R %U 