Using Function Approximation to Determine Neural Network Accuracy

2013 ◽  
Vol 2 (1) ◽  
pp. 89-98
Author(s):  
R.F. Wichman ◽  
J. Alexander
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Function Approximation ◽  
High Reliability ◽  
Back Propagation ◽  
Accurate Method ◽  
Back Propagation Neural Network ◽  
Degree Of Approximation ◽  
Feed Forward Back Propagation ◽  
Non Linear

Many, if not most, control processes demonstrate non-linear behavior in some portion of their operating range and the ability of neural networks to model non-linear dynamics makes them very appealing for control. Control of high reliability safety systems, and autonomous control in process or robotic applications, however, require accurate and consistent control and neural networks are only approximators of various functions so their degree of approximation becomes important. In this paper, the factors affecting the ability of a feed-forward back-propagation neural network to accurately approximate a non-linear function are explored. Compared to pattern recognition using a neural network for function approximation provides an easy and accurate method for determining the network's accuracy. In contrast to other techniques, we show that errors arising in function approximation or curve fitting are caused by the neural network itself rather than scatter in the data. A method is proposed that provides improvements in the accuracy achieved during training and resulting ability of the network to generalize after training. Binary input vectors provided a more accurate model than with scalar inputs and retraining using a small number of the outlier x,y pairs improved generalization.

scholarly journals A Framework for Prediction of Household Energy Consumption Using Feed Forward Back Propagation Neural Network

Technologies ◽  
2019 ◽  
Vol 7 (2) ◽  
pp. 30 ◽  
Author(s):  
Muhammad Fayaz ◽  
Habib Shah ◽  
Ali Aseere ◽  
Wali Mashwani ◽  
Abdul Shah
Keyword(s):  
Neural Network ◽  
Performance Evaluation ◽  
Energy Consumption ◽  
Data Acquisition ◽  
Residential Buildings ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Statistical Moments ◽  
Feed Forward ◽  
Feed Forward Back Propagation

Energy is considered the most costly and scarce resource, and demand for it is increasing daily. Globally, a significant amount of energy is consumed in residential buildings, i.e., 30–40% of total energy consumption. An active energy prediction system is highly desirable for efficient energy production and utilization. In this paper, we have proposed a methodology to predict short-term energy consumption in a residential building. The proposed methodology consisted of four different layers, namely data acquisition, preprocessing, prediction, and performance evaluation. For experimental analysis, real data collected from 4 multi-storied buildings situated in Seoul, South Korea, has been used. The collected data is provided as input to the data acquisition layer. In the pre-processing layer afterwards, several data cleaning and preprocessing schemes are applied to the input data for the removal of abnormalities. Preprocessing further consisted of two processes, namely the computation of statistical moments (mean, variance, skewness, and kurtosis) and data normalization. In the prediction layer, the feed forward back propagation neural network has been used on normalized data and data with statistical moments. In the performance evaluation layer, the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean squared error (RMSE) have been used to measure the performance of the proposed approach. The average values for data with statistical moments of MAE, MAPE, and RMSE are 4.3266, 11.9617, and 5.4625 respectively. These values of the statistical measures for data with statistical moments are less as compared to simple data and normalized data which indicates that the performance of the feed forward back propagation neural network (FFBPNN) on data with statistical moments is better when compared to simple data and normalized data.

scholarly journals Automatical Knowledge Representation of Logical Relations by Dynamical Neural Network

2017 ◽  
Vol 26 (4) ◽  
pp. 625-639 ◽  
Author(s):  
Gang Wang
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Knowledge Representation ◽  
Network Structure ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Ann Model ◽  
Functional Approximation ◽  
Dynamical Neural Network

AbstractCurrently, most artificial neural networks (ANNs) represent relations, such as back-propagation neural network, in the manner of functional approximation. This kind of ANN is good at representing the numeric relations or ratios between things. However, for representing logical relations, these ANNs have disadvantages because their representation is in the form of ratio. Therefore, to represent logical relations directly, we propose a novel ANN model called probabilistic logical dynamical neural network (PLDNN). Inhibitory links are introduced to connect exciting links rather than neurons so as to inhibit the connected exciting links conditionally to make them represent logical relations correctly. The probabilities are assigned to the weights of links to indicate the belief degree in logical relations under uncertain situations. Moreover, the network structure of PLDNN is less limited in topology than traditional ANNs, and it is dynamically built completely according to the data to make it adaptive. PLDNN uses both the weights of links and the interconnection structure to memorize more information. The model could be applied to represent logical relations as the complement to numeric ANNs.

Prediction of Human Responses to Dairy Odor Using an Electronic Nose and Neural Networks

2018 ◽  
Vol 61 (2) ◽  
pp. 399-409 ◽  
Author(s):  
Fangle Chang ◽  
Paul Heinemann
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Electronic Nose ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Sources Of Information ◽  
Independent Validation ◽  
Quantitative Measurements ◽  
Resilient Back Propagation ◽  
Human Assessment

Abstract. Odor emitted from dairy operations may cause negative reactions by farm neighbors. Identification and evaluation of such malodors is vital for better understanding of human response and methods for mitigating effects of odors. The human nose is a valuable tool for odor assessment, but using human panels can be costly and time-consuming, and human evaluation of odor is subjective. Sensing devices, such as an electronic nose, have been widely used to measure volatile emissions from different materials. The challenge, though, is connecting human assessment of odors with the quantitative measurements from instruments. In this work, a prediction system was designed and developed to use instruments to predict human assessment of odors from common dairy operations. The model targets are the human responses to odor samples evaluated using a general pleasantness scale ranging from -11 (extremely unpleasant) to +11 (extremely pleasant). The model inputs were the electronic nose measurements. Three different neural networks, a Levenberg-Marquardt back-propagation neural network (LMBNN), a scaled conjugate gradient back-propagation neural network (CGBNN), and a resilient back-propagation neural network (RPBNN), were applied to connect these two sources of information (human assessments and instrument measurements). The results showed that the LMBNN model can predict human assessments with accuracy as high as 78% within a 10% range and as high as 63% within a 5% range of the targets in independent validation. In addition, the LMBNN model performed with the best stability in both training and independent validation. Keywords: Animal production, Hedonic tone, Olfactometric models.

scholarly journals DEVELOPMENT OF BIOMETRIC METHODS AND INFORMATION SECURITY TOOLS

2021 ◽  
Vol 2 (336) ◽  
pp. 121-124
Author(s):  
T. Zh. Mazakov ◽  
D. N. Narynbekovna
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Face Recognition ◽  
Learning Algorithm ◽  
Back Propagation ◽  
Principal Component ◽  
Recognition System ◽  
Back Propagation Neural Network ◽  
Future Perspective ◽  
Data Set

Now a day’s security is a big issue, the whole world has been working on the face recognition techniques as face is used for the extraction of facial features. An analysis has been done of the commonly used face recognition techniques. This paper presents a system for the recognition of face for identification and verification purposes by using Principal Component Analysis (PCA) with Back Propagation Neural Networks (BPNN) and the implementation of face recognition system is done by using neural network. The use of neural network is to produce an output pattern from input pattern. This system for facial recognition is implemented in MATLAB using neural networks toolbox. Back propagation Neural Network is multi-layered network in which weights are fixed but adjustment of weights can be done on the basis of sigmoidal function. This algorithm is a learning algorithm to train input and output data set. It also calculates how the error changes when weights are increased or decreased. This paper consists of background and future perspective of face recognition techniques and how these techniques can be improved.

scholarly journals Isobaric vapour-liquid equilibrium calculations of binary systems using neural network

2004 ◽  
Vol 69 (8-9) ◽  
pp. 669-674 ◽  
Author(s):  
Mehmet Bilgin
Keyword(s):  
Neural Network ◽  
Binary Systems ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Liquid Equilibrium ◽  
Chemical Structures ◽  
Feed Forward Back Propagation ◽  
The Neural Network ◽  
Classical Models ◽  
Vle Data

A model on a feed forward back propagation neural network was employed to calculate the isobaric vapour?liquid equilibrium (VLE) data at 40, 66.67, and 101.32 ??0.02 kPa for the methylcyclohexane ? toluene and isopropanol ? methyl isobutyl ketone binary systems, which are composed of different chemical structures (cyclic, aromatic, alcohol and ketone) and do not show azeotrope behaviour. Half of the experimental VLE data only were assigned into the designed framework as training patterns in order to estimate the VLE data over the whole composition range at the mentioned pressures. The results were compared with the data calculated by the two classical models used in this field, the UNIFAC and Margules models. In all cases the deviations the experimental activity coefficients and those calculated by the neural network model (NNET) were lower than those obtained using the Margules and UNIFAC models.

FFBPNN-BASED HIGH DROWSINESS CLASSIFICATION USING EMG AND WPT

2020 ◽  
Vol 32 (03) ◽  
pp. 2050023 ◽  
Author(s):  
Mousa Kadhim Wali
Keyword(s):  
Neural Network ◽  
Wavelet Packet ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Road Accidents ◽  
Mean Values ◽  
Physiological Signal ◽  
Feed Forward Back Propagation ◽  
Emg Signal ◽  
Amplitude Level

The detection of drowsiness level is important because it is the main reason for fatal road accidents. Electromyography of the upper arm and shoulder is an important physiological signal affected by drivers’ drowsiness, in which its amplitude level and frequency band of the sleep-deprived case are different than those of the alert state. Therefore depending on electromyography (EMG), its drowsiness frequency (80–100[Formula: see text]Hz) was detected in order to determine high drowsiness state based on wavelet packet transform (WPT) which decomposes the EMG signal into its approximation and detail coefficients up to level 4 using db2, db7, sym5 and coif5 wavelets. In this research after extraction, the two higher order statistical features, kurtosis and skewness, are computed from 3[Formula: see text]s window of the three EMG channels, and analysis of variance test is used to check whether their mean values are different for the different classes as both [Formula: see text]-values are less than 0.005 under db2 wavelet. Therefore, they were supplied to feed forward back propagation neural network (FFBPNN) as this type of neural network is used for distinguishing and classification purposes for different objects. They obtained an accuracy of 75% for detecting high levels among other levels of normal and low drowsiness with an average sensitivity of 78.63% and specificity of 75.97% because the spectrum of the EMG alert (non-drowsiness) signal of 80–100 Hz is different from that of drowsy 80–90[Formula: see text]Hz and high drowsy 78–95[Formula: see text]Hz signals.

Sign in / Sign up

Export Citation Format

Share Document