“…[26] This technique was designed to approach the second-order training speed without the need for computing the Hessian matrix. [26,27] Considering the effectiveness and efficiency of Newton's method, LMNN aims at shifting to Newton's method as fast as possible. As one of the faster training algorithms, RPNN eliminates the effects caused by partial derivatives often associated with multi-layered networks trained with sigmoid functions.…”
Due to the inherent multiple response characteristics in many biological and separation processes, parameter optimization and modelling is usually a daunting task. The integration of Deng's grey incidence model (GRA) and Taguchi optimization (TM) therefore helps in transforming multiple quality characteristics into a single response presented as the grey relational grade (GRG). This was applied to optimize the multiple quality response characteristics in the maceration-assisted extraction of African cucumber leaves. Two responses and five design factors were selected with L 16 (2 5 ) layout using signal-to-noise ratio as a point prediction feature. Under the optimized conditions, the optimum total phenolic content and antioxidant capacity of 0.8569 mg/ml gallic acid equivalence and 0.9259 mg/ml were achieved, respectively. The mass ratio was the highest contributor (38.2%), whereas the maceration time presented the least contribution (9.8%) to the cumulative response grade (GRG). In the neural network analysis, three models were deployed: Levenberg Marquardt backpropagation neural network (LMNN), gradient descent with adaptive learning rate neural network (GDALRNN), and the resilient back-propagation neural network (RPNN). A better prediction of hold-out data was achieved with the GDALRNN model, generating lesser absolute deviation error (MAD GDALRNN = 0.099), root mean square error (RMSE GDALRNN = 0.1033), relative mean bias error (rMBE GDALRNN = À 0.24), and highest computational time (CT GDALRNN = 8.8), which is expected of an effective model. Based on the GRG and the signal-to-noise ratio, the optimum conditions and the neural network model succinctly provided a benchmark for future assessment of complex relationship among extraction variables, which could form the basis for a potential future scale-up applications.
“…[26] This technique was designed to approach the second-order training speed without the need for computing the Hessian matrix. [26,27] Considering the effectiveness and efficiency of Newton's method, LMNN aims at shifting to Newton's method as fast as possible. As one of the faster training algorithms, RPNN eliminates the effects caused by partial derivatives often associated with multi-layered networks trained with sigmoid functions.…”
Due to the inherent multiple response characteristics in many biological and separation processes, parameter optimization and modelling is usually a daunting task. The integration of Deng's grey incidence model (GRA) and Taguchi optimization (TM) therefore helps in transforming multiple quality characteristics into a single response presented as the grey relational grade (GRG). This was applied to optimize the multiple quality response characteristics in the maceration-assisted extraction of African cucumber leaves. Two responses and five design factors were selected with L 16 (2 5 ) layout using signal-to-noise ratio as a point prediction feature. Under the optimized conditions, the optimum total phenolic content and antioxidant capacity of 0.8569 mg/ml gallic acid equivalence and 0.9259 mg/ml were achieved, respectively. The mass ratio was the highest contributor (38.2%), whereas the maceration time presented the least contribution (9.8%) to the cumulative response grade (GRG). In the neural network analysis, three models were deployed: Levenberg Marquardt backpropagation neural network (LMNN), gradient descent with adaptive learning rate neural network (GDALRNN), and the resilient back-propagation neural network (RPNN). A better prediction of hold-out data was achieved with the GDALRNN model, generating lesser absolute deviation error (MAD GDALRNN = 0.099), root mean square error (RMSE GDALRNN = 0.1033), relative mean bias error (rMBE GDALRNN = À 0.24), and highest computational time (CT GDALRNN = 8.8), which is expected of an effective model. Based on the GRG and the signal-to-noise ratio, the optimum conditions and the neural network model succinctly provided a benchmark for future assessment of complex relationship among extraction variables, which could form the basis for a potential future scale-up applications.
“…As an intelligent technique, the multilayer perceptron has been widely used for optimisation, modelling, prediction and function approximation purposes [105]. However, it has also been successfully applied to a variety of pattern recognition and classification problems [54,55]. Such applications include disease recognition [77], physiological analysis and modeling [46], cancer detection and classification [47], modelling of heart disease recognition [106], diagnosis of coronary artery disease [49], and other related studies [52,89].…”
<span lang="EN-US">Intelligence study is one of keystone to distinguish individual differences in cognitive psychology. Conventional psychometric tests are limited in terms of assessment time, and existence of biasness issues. Apart from that, there is still lack in knowledge to classify IQ based on EEG signals and intelligent signal processing (ISP) technique. ISP purpose is to extract as much information as possible from signal and noise data using learning and/or other smart techniques. Therefore, as a first attempt in classifying IQ feature via scientific approach, it is important to identify a relevant technique with prominent paradigm that is suitable for this area of application. Thus, this article reviews several ISP approaches to provide consolidated source of information. This in particular focuses on prominent paradigm that suitable for pattern classification in biomedical area. The review leads to selection of ANN since it has been widely implemented for pattern classification in biomedical engineering. </span>
“…Όσον αφορά την επιτάχυνση των διαδικασιών εκπαίδευσης, συχνά χρησιµοποιούνται µεταβλητοί ρυθµοί µάθησης (adaptive learning rate), οι οποίοι προσαρµόζονται ανάλογα µε τις µεταβολές της συνάρτησης κόστους. Παράλληλα, έχουν αναπτυχθεί διάφορες αλγεβρικές τεχνικές γρήγορης εκπαίδευσης (faster training), όπως είναι οι αλγόριθµοι Quasi-Newton και Levenberg-Marquard [254]- [255], [263] και η µέθοδος "resilient propagation" (RPROP) [270]. Η τελευταία χρησιµοποιήθηκε εκτενώς για τις ανάγκες της παρούσης διατριβής, παρέχοντας µεγάλες ταχύτητες εκπαίδευσης και οµαλές διακυµάνσεις κατά την «πτώση» της καµπύλης µέσου σφάλµατος [216], [218].…”
Section: αναγνώριση προτύπων µε χρήση νευρωνικών δικτύωνunclassified
“…Η τελευταία χρησιµοποιήθηκε εκτενώς για τις ανάγκες της παρούσης διατριβής, παρέχοντας µεγάλες ταχύτητες εκπαίδευσης και οµαλές διακυµάνσεις κατά την «πτώση» της καµπύλης µέσου σφάλµατος [216], [218]. Στη µέθοδο RPROP, οι συντελεστές µεταβάλλονται κατά σταθερές ποσότητες A ℓi,ℓj (tr) , µε βάση µόνο το πρόσηµο και όχι το µέτρο της κλίσης ∂J/∂W [254], [263], [270]:…”
Section: αναγνώριση προτύπων µε χρήση νευρωνικών δικτύωνunclassified
Εννοείται ο πόνος που δεν οφείλεται σε κάποια παθολογική κατάσταση του πεπτικού συστήµατος ή κάποιου άλλου γειτονικού οργάνου, αλλά σε λειτουργικά ζητήµατα κινητικότητας. * Τα τεστ, που διεξάγονται σε δείγµατα υποκειµένων (εθελοντές) πριν την έγκριση κυκλοφορίας των φαρµάκων, βασίζονται αποκλειστικά σε αξιολόγηση σχετικών ερωτηµατολογίων. † Η µετεγχειρητική λειτουργία του εντέρου κρίνεται ιδιαίτερα σηµαντική στην περαιτέρω πορεία ανάρρωσης του ασθενή.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.