9 Hits in 1.8 sec

Impact of Air Pollution on Respiratory Diseases: Correlation and Classification by Multivariate Data Analysis

M. Deepa, M. Rajalakshmi, R. Nedunchezhian
2017 Data-Enabled Discovery and Applications  
In this paper, an extreme learning machine (ELM) based on statistically controlled activation weight initialization is used to learn and measure the correlation between PM 2.5 and lung-related problems  ...  A statistically controlled activation weight initialization (SCAWI) [14] method is used for initializing input weights and bias.  ...  Here, an ELM with dual-activation function [14] for hidden node and a statistically controlled activation weight initialization [4] for hidden node parameters are proposed for improving the accuracy  ... 
doi:10.1007/s41688-017-0004-z fatcat:ypvvorfb25dx5bu6b5ez2xsg4e

Solving the linear interval tolerance problem for weight initialization of neural networks

S.P. Adam, D.A. Karras, G.D. Magoulas, M.N. Vrahatis
2014 Neural Networks  
Determining good initial conditions for an algorithm used to train a neural network is considered a parameter estimation problem dealing with uncertainty about the initial weights.  ...  George Dimakopoulos, statistical analyst at the Technological Educational Institute of Epirus, for his helpful comments regarding ANOVA.  ...  Their method is called statistically controlled activation weight initialization (SCAWI). They determine the maximum magnitude of the weights through statistical analysis.  ... 
doi:10.1016/j.neunet.2014.02.006 pmid:24637071 fatcat:5x63io6omjh3zju3uvtwtu7b5u

Training Process Reduction Based On Potential Weights Linear Analysis To Accelarate Back Propagation Network [article]

Roya Asadi, Norwati Mustapha, Nasir Sulaiman
2009 arXiv   pre-print
Currently, data pre-processing such as dimension reduction input values and pre-training are the contributing factors in developing efficient techniques for reducing training time with high accuracy and initialization  ...  After these phases, dimension of input values matrix will be reduced by using real potential weights.  ...  The disadvantage of the method of MinMax is usage of initialization with random numbers which create critical in training. 2) SCAWI: The method called Statistically Controlled Activation Weight Initialization  ... 
arXiv:0908.1453v1 fatcat:nalgcr6v3ncsboa6dzyzt266ae


Farhana Kausar, Dr. Aishwarya P., Dr. Gopal Krishna Shyam
2021 Information Technology in Industry  
The newly proposed paradigm Xavier Weight Initialization less error in comparison to the traditional approach of Uniform and Gaussian Weight initialization (Random Initialization).  ...  Assuming all the above condition, it decided to initialize the neural network training by different weight initialization techniques.  ...  Mentioned two formula is used to calculate the values of weight A technique known as statistically regulated activation weight initialization (SCAWI) to and optimum initial weights [13] .  ... 
doi:10.17762/itii.v9i1.290 fatcat:6jytk3ditve5ldeelqiavuh4hq

Towards an optimal set of initial weights for a Deep Neural Network architecture

Abdelhalim Saadi, Hacene Belhadef
2019 Neural Network World  
The neural network is first trained on reduced data represented by a set of centroids of the original data points, and then the learned weights are used to initialize a second training phase of the neural  ...  In [16] , a weight initialization method called statistically controlled activation weight initialization (SCAWI) was developed.  ...  The initial values of the weights for a hidden layer i should be uniformly sampled from a symmetric interval that depends on an activation function.  ... 
doi:10.14311/nnw.2019.29.025 fatcat:dlfs7u72kfaptbycthqsb6nkru

Multi-Disease Classification Model using Strassen's Half of Threshold (SHoT) Training Algorithm in Healthcare Sector

Manjula Devi Ramasamy, Keerthika Periasamy, Lalitha Krishnasamy, Rajesh Kumar Dhanaraj, Seifedine Kadry, Yunyoung Nam
2021 IEEE Access  
Nguyen and Widrow [8] assign a fraction of the intended response range to each hidden node, and Drago and Ridella [9] utilize a technique called statistically controlled activation weight initialization  ...  Initial weights have been demonstrated to affect the BPN technique. [7] . In most cases, modest random numbers are chosen as the NN's initial weights.  ... 
doi:10.1109/access.2021.3103746 fatcat:2ncaqtqxyzbd3jiczvd5ufwete

Initialization and Estimation of Weights and Bias using Bayesian Technique

Oryiema Robert, David Angwenyi, Nyogesa Achiles, Jacob On'gala
2022 Asian Journal of Probability and Statistics  
The research applies the new FoEEF Filter to initializes and estimates the weights and bias of artificial neural network.  ...  Researches on artificial neural network models have shown that the method used to initialize and estimate weights and bias always determines the rate at which the network will converge and how efficient  ...  Drago et al [11] applied statistically controlled activation weight initialization (SCAWI) method to determine initial weights of their model.  ... 
doi:10.9734/ajpas/2022/v17i230420 fatcat:pzogvfmjb5czxh5rxcxit7pnfq

Autonomous Design of Artificial Neural Networks by Neurex

François Michaud, Ruben Gonzalez Rubio
1996 Neural Computation  
SCAWI: An algorithm for weight initialization of a sigmoidal neural network. In Artificial Neural Networks 2, pp. 983-986. Elsevier Science. Fahlman, S. E. 1988.  ...  Learning rate increment and random initial weights incre- 7f. = 2 is used.  ... 
doi:10.1162/neco.1996.8.8.1767 pmid:8888617 fatcat:3fibl7ljcnf7pjrs5icyd2t2xy

Modelling of turbulent combustion using the Rate-Controlled Constrained Equilibrium (RCCE) - Arti cial Neural Networks (ANNs) approach

Athanasios Chatzopoulos, Stelios Rigopoulos, Engineering And Physical Sciences Research Council
This task is performed by the Rate-Controlled Constrained Equilibrium (RCCE) reduction method.  ...  Weights are uniformly distributed over the interval of [−A, A] with A = Kim and Ra method Statistically Controlled Activation Weight Initialization (SCAWI) √ 1+0.3N in for neurons of hidden and  ...  This is the type of weight initialization used by most 2. Fan-in random initialisation [139] .  ... 
doi:10.25560/30782 fatcat:bnibjwitpvd6vnukgiour7xzxa