Configuring and Optimizing the Back-Propagation Network

Maren, Alianna J.; Jones, Dan; Franklin, Stanley P.

doi:10.1016/b978-0-12-546090-3.50019-x

Cited by 20 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another crucial aspect in ANN modelling is the number of neurons in the first hidden layer because it is responsible for defining the main features of the anticipated output curve such as the order of the curve and the segments at which the curve changes the direction of its slope. Indeed, a wrong choice of number of neuron in the hidden layer value could determine a low ANN performance, wellknown as weight/bias variance dilemma [49].…”

Section: Depth Of Cut (Mm)mentioning

confidence: 99%

“…To overcome this drawback, it is possible to apply some heuristics that allow finding a lower and upper bound for the number of neuron in the hidden layer [49]. These approaches prescribe the better way to fix the number of neurons into the hidden layers dependent on the number of the input (NIN) and the output (NON) neurons, properly decreased or increased according to a constant.…”

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

“…From preliminary investigation, the lowest testing error was found with the heuristic proposed by Kolmogorov [49], and so the numbers of hidden layer neurons were kept 9.…”

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

“…the Freeman-Skapura initialization, where the weights are initialized randomly in the range [−0.5, 0.5]; the Nguyen-Widrow (NWI) and the Fan-In-based initialization where, vice versa, the NN morphology has been taken into account firstly [46]. It was observed that NWI allow reducing the computational time associated to the NN training phase since both optimize the initial values of weights, reducing the convergence time of the algorithm [46,49]. As a result in the present investigation, NWI has been considered for the weight initialization.…”

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

See 3 more Smart Citations

Prediction of depth of cut for single-pass laser micro-milling process using semi-analytical, ANN and GP approaches

Desai

Shaikh

2011

Int J Adv Manuf Technol

View full text Add to dashboard Cite

The present study is aimed to investigate micromilling performance of thermoplastics with different parameters, namely laser beam absorptivity, latent heat of vaporization, laser power and cutting speed. The 25-W CO 2 (CW) laser engraving machine is used for the investigation. In total 50 different combinations of laser power and cutting speed with four categories of thermoplastics, namely poly-methyl-methacrylate, poly-propylene, acrylonitrile butadiene styrene and nylon 6, are used in this study. Experimental results suggest that laser beam absorptivity, cutting power and cutting speed are the major influencing parameters on depth of cut. Theoretical model for the prediction of depth of cut in terms of material properties, cutting power and cutting speed has been proposed. Two correction parameters have been introduced in this analysis using non-linear regression method to improve the theoretical model. Comparison has been made between prediction capabilities of theoretical model, model based on multigene genetic programming and artificial neural network. The comparison clearly indicates that all the three models provide accurate prediction of depth of cut. The details of experimentation, model development, testing and the performance comparison are presented in this paper.

show abstract

Section: Depth Of Cut (Mm)mentioning

confidence: 99%

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

“…From preliminary investigation, the lowest testing error was found with the heuristic proposed by Kolmogorov [49], and so the numbers of hidden layer neurons were kept 9.…”

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

Section: Depth Of Cut (Mm) Vs Cutting Speed(mm/sec)mentioning

confidence: 99%

See 2 more Smart Citations

Prediction of depth of cut for single-pass laser micro-milling process using semi-analytical, ANN and GP approaches

Desai

Shaikh

2011

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…For selecting the eligible number of HN in the single hidden layer, a range in which the number of neurons varies was considered taking into account some heuristic methods to define the upper and the lower limit for the HN as a function of IN and ON. Specifically, according to the literature [23,24], the lower bound (MTI) was defined as reported in Equation (2).…”

Section: Ann-pso Model 221 Ann Designmentioning

confidence: 99%

A Comparison between Finite Element Model (FEM) Simulation and an Integrated Artificial Neural Network (ANN)-Particle Swarm Optimization (PSO) Approach to Forecast Performances of Micro Electro Discharge Machining (Micro-EDM) Drilling

et al. 2021

View full text Add to dashboard Cite

Artificial Neural Network (ANN), together with a Particle Swarm Optimization (PSO) and Finite Element Model (FEM), was used to forecast the process performances for the Micro Electrical Discharge Machining (micro-EDM) drilling process. The integrated ANN-PSO methodology has a double direction functionality, responding to different industrial needs. It allows to optimize the process parameters as a function of the required performances and, at the same time, it allows to forecast the process performances fixing the process parameters. The functionality is strictly related to the input and/or output fixed in the model. The FEM model was based on the capacity of modeling the removal process through the mesh element deletion, simulating electrical discharges through a proper heat-flux. This paper compares these prevision models, relating the expected results with the experimental data. In general, the results show that the integrated ANN-PSO methodology is more accurate in the performance previsions. Furthermore, the ANN-PSO model is faster and easier to apply, but it requires a large amount of historical data for the ANN training. On the contrary, the FEM is more complex to set up, since many physical and thermal characteristics of the materials are necessary, and a great deal of time is required for a single simulation.

show abstract

Classification of normal and abnormal electrogastrograms using multilayer feedforward neural networks

Lin

Maris²,

Hermans³

et al. 1997

Med. Biol. Eng. Comput.

View full text Add to dashboard Cite

A neural network approach is proposed for the automated classification of the normal and abnormal EGG. Two learning algorithms, the quasi-Newton and the scaled conjugate gradient method for the multilayer feedforward neural networks (MFNN), are introduced and compared with the error backpropagation algorithm. The configurations of the MFNN are determined by experiment. The raw EGG data, its power spectral data, and its autoregressive moving average (ARMA) modelling parameters are used as the input to the MFNN and compared with each other. Three indexes (the percent correct, sum-squared error and complexity per iteration) are used to evaluate the performance of each learning algorithm. The results show that the scaled conjugate gradient algorithm performs best, in that it is robust and provides a super-linear convergence rate. The power spectral representation and the ARMA modelling parameters of the EGG are found to be better types of the input to the network for this specific application, both yielding a percent correctness of 95% on the test set. Although the results are focused on the classification of the EGG, this paper should provide useful information for the classification of other biomedical signals.

show abstract

Configuring and Optimizing the Back-Propagation Network

Cited by 20 publications

References 18 publications

Prediction of depth of cut for single-pass laser micro-milling process using semi-analytical, ANN and GP approaches

Prediction of depth of cut for single-pass laser micro-milling process using semi-analytical, ANN and GP approaches

A Comparison between Finite Element Model (FEM) Simulation and an Integrated Artificial Neural Network (ANN)-Particle Swarm Optimization (PSO) Approach to Forecast Performances of Micro Electro Discharge Machining (Micro-EDM) Drilling

Classification of normal and abnormal electrogastrograms using multilayer feedforward neural networks

Contact Info

Product

Resources

About