Bayesian Regularization of Neural Networks

2008 ◽  
pp. 23-42 ◽  
Author(s):  
Frank Burden ◽  
Dave Winkler
Keyword(s):  
Neural Networks ◽  
Bayesian Regularization

Large-scale Groundwater Simulation using Artificial Neural Networks in the Danube River Basin

2020 ◽  
Author(s):  
Illias Landros ◽  
Ioannis Trichakis ◽  
Emmanouil Varouchakis ◽  
George P. Karatzas
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
River Basin ◽  
Danube River ◽  
Bayesian Regularization ◽  
Data Set ◽  
The Danube River ◽  
Precipitation Temperature ◽  
Artificial Neural ◽  
Danube River Basin

<p>In recent years, Artificial Neural Networks (ANNs) have proven their merit in being able to simulate the changes in groundwater levels, using as inputs other parameters of the water budget, e.g. precipitation, temperature, etc.. In this study, ANNs have been used to simulate hydraulic head in a large number of wells throughout the Danube River Basin, taking as inputs, precipitation, temperature, and evapotranspiration data in the region. Different ANN architectures have been examined, to minimize the simulation error of the testing data-set. Among the different training algorithms, Levenberg-Marquardt and Bayesian Regularization are used to train the ANNs, while the different activation functions of the neurons that were deployed include tangent sigmoid, logarithmic sigmoid and linear. The initial application comprised of data from 128 wells between 1 January 2000 and 31 October 2014. The best performance was achieved by the algorithm Bayesian Regularization with a error of the order  based on all observation wells. A second application, compared the results of the first one, with the results of an ANN used to simulate a single well. The pros and cons of the two approaches, and the synergies of using both of them is further discussed in order to distinguish the differences, and guide researchers in the field for further applications.</p>

scholarly journals Self-organizing Maps and Bayesian Regularized Neural Network for Analyzing Gasoline and Diesel Price Drifts

2022 ◽  
Vol 15 (1) ◽  
Author(s):  
R. Sujatha ◽  
Jyotir Moy Chatterjee ◽  
Ishaani Priyadarshini ◽  
Aboul Ella Hassanien ◽  
Abd Allah A. Mousa ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Conjugate Gradient ◽  
Bayesian Regularization ◽  
Training Algorithms ◽  
Fuel Price ◽  
Self Organizing Maps ◽  
Scaled Conjugate Gradient ◽  
Indirect Impacts ◽  
Self Organizing

AbstractAny nation’s growth depends on the trend of the price of fuel. The fuel price drifts have both direct and indirect impacts on a nation’s economy. Nation’s growth will be hampered due to the higher level of inflation prevailing in the oil industry. This paper proposed a method of analyzing Gasoline and Diesel Price Drifts based on Self-organizing Maps and Bayesian regularized neural networks. The US gasoline and diesel price timeline dataset is used to validate the proposed approach. In the dataset, all grades, regular, medium, and premium with conventional, reformulated, all formulation of gasoline combinations, and diesel pricing per gallon weekly from 1995 to January 2021, are considered. For the data visualization purpose, we have used self-organizing maps and analyzed them with a neural network algorithm. The nonlinear autoregressive neural network is adopted because of the time series dataset. Three training algorithms are adopted to train the neural networks: Levenberg-Marquard, scaled conjugate gradient, and Bayesian regularization. The results are hopeful and reveal the robustness of the proposed model. In the proposed approach, we have found Levenberg-Marquard error falls from − 0.1074 to 0.1424, scaled conjugate gradient error falls from − 0.1476 to 0.1618, and similarly, Bayesian regularization error falls in − 0.09854 to 0.09871, which showed that out of the three approaches considered, the Bayesian regularization gives better results.

A Digital Pre-Distortion Scheme for CMMB Repeater System Based on Dynamic Neural Networks

2013 ◽  
Vol 336-338 ◽  
pp. 1738-1743
Author(s):  
Ze Biao Lin ◽  
Chun Hui Huang
Keyword(s):  
Neural Networks ◽  
Computer Simulation ◽  
Memory Effect ◽  
Orthogonal Frequency Division Multiplexing ◽  
Bayesian Regularization ◽  
Frequency Division Multiplexing ◽  
Frequency Division ◽  
Dynamic Neural Networks ◽  
Adjacent Channel ◽  
First Time

CMMB technology employs Orthogonal Frequency Division Multiplexing (OFDM) modulation with 4096 subcarriers for 8MHz bandwidth mode, HPA in the CMMB repeater will cause significant distortion and spectral extension. To compensate HPA nonlinearity and memory effect in a CMMB repeater system, this paper proposed a pre-distortion scheme based on RNN and Bayesian Regularization for the first time. Computer simulation is used to confirm the validation of the proposed scheme and an adjacent channel leakage ratio (ACLR) improvement of 20 dB is obtained in the paper.

scholarly journals Non-linear carbon dioxide determination using infrared gas sensors and neural networks with Bayesian regularization

2009 ◽  
Vol 136 (1) ◽  
pp. 242-247 ◽  
Author(s):  
King-Tong Lau ◽  
Weimin Guo ◽  
Breda Kiernan ◽  
Conor Slater ◽  
Dermot Diamond
Keyword(s):  
Carbon Dioxide ◽  
Neural Networks ◽  
Gas Sensors ◽  
Bayesian Regularization ◽  
Non Linear
Sign in / Sign up

Export Citation Format

Share Document