Comparative study of various training algorithms of artificial neural network

Jaiswal, Pragati; Gupta, Nikhil; Ambikapathy, A.

doi:10.1109/icacccn.2018.8748660

Cited by 69 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A typical numerical optimization approach can be used to optimize the performance function of multilayer feed-forward networks during training [29]. The various training algorithms that are available in the Deep Learning Toolbox software and that use gradient-or Jacobian-based methods are explained in the following section [30]. Various methods are available to stop training.…”

Section: Training Algorithmsmentioning

confidence: 99%

Performance analysis of various training algorithms of deep learning based controller

2023

View full text Add to dashboard Cite

Advances in artificial neural networks (ANN), specifically deep learning (DL), have widened the application domain of process control. DL algorithms and models have become quite common these days. The training algorithm is the most important part of an ANN that affects the performance of the controller. Training algorithms optimize the weights and biases of the ANN according to the input-output patterns. In this paper, the performance of different training algorithms was evaluated, analysed, and compared in a feed-forward backpropagation architecture. The training algorithms were simulated on MATLAB R2021b with license number 1075356. Training data were generated using two benchmark problems of the process control system. The performance, gradient, training error, validation error, testing error, and regression of the different training algorithms were obtained and analysed. The data shows that the Levenberg-Marquardt (LM) algorithm produced the best validation performance with a value of 2.669*10−14 at 2000 epochs, while ‘traingd’ and ‘traingdm’ algorithms did not improve beyond their initial values. The LM algorithm tends to produce better results than other algorithms. These results indicate that the LM backpropagation best suits these types of benchmark problems. The results also suggest that the choice of training algorithm can significantly impact the performance of a neural network.

show abstract

Section: Training Algorithmsmentioning

confidence: 99%

Performance analysis of various training algorithms of deep learning based controller

2023

View full text Add to dashboard Cite

show abstract

“…The algorithm uses the advantage of both methods and chooses the one according to closeness to the optimal value. The iterative training algorithm makes LM faster than other training algorithms [35].…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Artificial Neural Network Based Prediction of Long-Term Electric Field Strength Level Emitted by 2G/3G/4G Base Station

Engiz

2023

Applied Sciences

View full text Add to dashboard Cite

Accurate predictions of radio frequency electromagnetic field (RF-EMF) levels can help implement measures to reduce exposure and check regulatory compliance. Therefore, this study aims to predict the RF-EMF levels in the medium using an artificial neural network (ANN). The work was conducted at Ondokuz Mayis University, Kurupelit Campus, where the measurement location has line-of-sight to the base stations. Band selective measurements were also performed to assess the contribution of 2G/3G/4G services to the total RF-EMF level, which was found to be the highest among all services within the total band. Long-term RF-EMF measurements were carried out for 35 days within the frequencies of 100 kHz to 3 GHz. Then, an ANN model with Levenberg–Marquardt (LM) and Bayesian Regulation (BR) algorithms was proposed, which utilized inputs from real-time RF-EMF measurements. The performance of the models was assessed in terms of mean squared error (MSE) and regression performance. The average MSE and regression performances of the models were similar, with the lowest testing MSEs of 2.78 × 10−3 and 3.76 × 10−3 for LM and BR methods, respectively. The analysis of the models showed that the proposed models help to predict the RF-EMF level in the medium with up to 99% accuracy.

show abstract

“…Finally, to train (or optimize) the ANN in order to update the weights and to minimize the approximation error (17) (more, precisely, its error norm, the Euclidean distance ( 20)), a training algorithm must be chosen [68]. Very often, the Gradient descent method is applied to update the weights by using the negative and by some 0 < η < 1 scaled gradient of the error norm [63], Section 10.4.4, i.e.,…”

Section: Preliminaries: Data Sets Performance Measures and Training A...mentioning

confidence: 99%

Artificial Neural Network Based Optimal Feedforward Torque Control of Interior Permanent Magnet Synchronous Machines: A Feasibility Study and Comparison with the State-of-the-Art

2022

View full text Add to dashboard Cite

A novel Artificial Neural Network (ANN) Based Optimal Feedforward Torque Control (OFTC) strategy is proposed which, after proper ANN design, training and validation, allows to analytically compute the optimal reference currents (minimizing copper and iron losses) for Interior Permanent Magnet Synchronous Machines (IPMSMs) with highly operating point dependent nonlinear electric and magnetic characteristics. In contrast to conventional OFTC, which either utilizes large look-up tables (LUTs; with more than three input parameters) or computes the optimal reference currents numerically or analytically but iteratively (due to the necessary online linearization), the proposed ANN-based OFTC strategy does not require iterations nor a decision tree to find the optimal operation strategy such as e.g., Maximum Torque per Losses (MTPL), Maximum Current (MC) or Field Weakening (FW). Therefore, it is (much) faster and easier to implement while (i) still machine nonlinearities and nonidealities such as e.g., magnetic cross-coupling and saturation and speed-dependent iron losses can be considered and (ii) very accurate optimal reference currents are obtained. Comprehensive simulation results for a real and highly nonlinear IPMSM clearly show these benefits of the proposed ANN-based OFTC approach compared to conventional OFTC strategies using LUT-based, numerical or analytical computation of the reference currents.

show abstract

Comparative study of various training algorithms of artificial neural network

Cited by 69 publications

References 8 publications

Performance analysis of various training algorithms of deep learning based controller

Performance analysis of various training algorithms of deep learning based controller

Artificial Neural Network Based Prediction of Long-Term Electric Field Strength Level Emitted by 2G/3G/4G Base Station

Artificial Neural Network Based Optimal Feedforward Torque Control of Interior Permanent Magnet Synchronous Machines: A Feasibility Study and Comparison with the State-of-the-Art

Contact Info

Product

Resources

About