Search results for: hardy cross networks accuracy

Authors: Ahmed Emad Ahmed, Zeyad Ahmed Hussein, Mohamed Salama Afifi, Ahmed Mohammed Eid

Abstract:

Water has a great importance in life. In order to deliver water from resources to the users, many procedures should be taken by the water engineers. One of the main procedures to deliver water to the community is by designing pressurizer pipe networks for water. The main aim of this work is to calculate the water demand of a small town and then design a simple water network to distribute water resources among the town with the smallest losses. Literature has been mentioned to cover the main point related to water distribution. Moreover, the methodology has introduced two approaches to solve the research problem, one by the iterative method of Hardy-cross and the other by water software Pipe Flow. The results have introduced two main designs to satisfy the same research requirements. Finally, the researchers have concluded that the use of water software provides more abilities and options for water engineers.

Keywords: looping pipe networks, hardy cross networks accuracy, relative error of hardy cross method

Procedia PDF Downloads 120

Authors: Nazmul Shahadat, Anthony S. Maida

Abstract:

In recent years, hypercomplex inspired neural networks improved deep CNN architectures due to their ability to share weights across input channels and thus improve cohesiveness of representations within the layers. The work described herein studies the effect of replacing existing layers in an Axial Attention ResNet with their quaternion variants that use cross-channel weight sharing to assess the effect on image classification. We expect the quaternion enhancements to produce improved feature maps with more interlinked representations. We experiment with the stem of the network, the bottleneck layer, and the fully connected backend by replacing them with quaternion versions. These modifications lead to novel architectures which yield improved accuracy performance on the ImageNet300k classification dataset. Our baseline networks for comparison were the original real-valued ResNet, the original quaternion-valued ResNet, and the Axial Attention ResNet. Since improvement was observed regardless of which part of the network was modified, there is a promise that this technique may be generally useful in improving classification accuracy for a large class of networks.

Keywords: axial attention, representational networks, weight sharing, cross-channel correlations, quaternion-enhanced axial attention, deep networks

Procedia PDF Downloads 42

Authors: Wedad Albalawi

Abstract:

The mathematical inequalities have been the core of mathematical study and used in almost all branches of mathematics as well in various areas of science and engineering. The inequalities by Hardy, Littlewood and Polya were the first significant composition of several science. This work presents fundamental ideas, results and techniques and it has had much influence on research in various branches of analysis. Since 1934, various inequalities have been produced and studied in the literature. Furthermore, some inequalities have been formulated by some operators; in 1989, weighted Hardy inequalities have been obtained for integration operators. Then, they obtained weighted estimates for Steklov operators that were used in the solution of the Cauchy problem for the wave equation. They were improved upon in 2011 to include the boundedness of integral operators from the weighted Sobolev space to the weighted Lebesgue space. Some inequalities have been demonstrated and improved using the Hardy–Steklov operator. Recently, a lot of integral inequalities have been improved by differential operators. Hardy inequality has been one of the tools that is used to consider integrity solutions of differential equations. Then dynamic inequalities of Hardy and Coposon have been extended and improved by various integral operators. These inequalities would be interesting to apply in different fields of mathematics (functional spaces, partial differential equations, mathematical modeling). Some inequalities have been appeared involving Copson and Hardy inequalities on time scales to obtain new special version of them. A time scale is defined as a closed subset contains real numbers. Then the inequalities of time scales version have received a lot of attention and has had a major field in both pure and applied mathematics. There are many applications of dynamic equations on time scales to quantum mechanics, electrical engineering, neural networks, heat transfer, combinatorics, and population dynamics. This study focuses on double integrals to obtain new time-scale inequalities of Copson driven by Steklov operator. They will be applied in the solution of the Cauchy problem for the wave equation. The proof can be done by introducing restriction on the operator in several cases. In addition, the obtained inequalities done by using some concepts in time scale version such as time scales calculus, theorem of Fubini and the inequality of H¨older.

Keywords: time scales, inequality of Hardy, inequality of Coposon, Steklov operator

Procedia PDF Downloads 39

Authors: Wedad Albalawi

Abstract:

The mathematical inequalities have been the core of mathematical study and used in almost all branches of mathematics as well in various areas of science and engineering. The inequalities by Hardy, Littlewood and Polya were the first significant composition of several science. This work presents fundamental ideas, results and techniques, and it has had much influence on research in various branches of analysis. Since 1934, various inequalities have been produced and studied in the literature. Furthermore, some inequalities have been formulated by some operators; in 1989, weighted Hardy inequalities have been obtained for integration operators. Then, they obtained weighted estimates for Steklov operators that were used in the solution of the Cauchy problem for the wave equation. They were improved upon in 2011 to include the boundedness of integral operators from the weighted Sobolev space to the weighted Lebesgue space. Some inequalities have been demonstrated and improved using the Hardy–Steklov operator. Recently, a lot of integral inequalities have been improved by differential operators. Hardy inequality has been one of the tools that is used to consider integrity solutions of differential equations. Then, dynamic inequalities of Hardy and Coposon have been extended and improved by various integral operators. These inequalities would be interesting to apply in different fields of mathematics (functional spaces, partial differential equations, mathematical modeling). Some inequalities have been appeared involving Copson and Hardy inequalities on time scales to obtain new special version of them. A time scale is an arbitrary nonempty closed subset of the real numbers. Then, the dynamic inequalities on time scales have received a lot of attention in the literature and has become a major field in pure and applied mathematics. There are many applications of dynamic equations on time scales to quantum mechanics, electrical engineering, neural networks, heat transfer, combinatorics, and population dynamics. This study focuses on Hardy and Coposon inequalities, using Steklov operator on time scale in double integrals to obtain special cases of time-scale inequalities of Hardy and Copson on high dimensions. The advantage of this study is that it uses the one-dimensional classical Hardy inequality to obtain higher dimensional on time scale versions that will be applied in the solution of the Cauchy problem for the wave equation. In addition, the obtained inequalities have various applications involving discontinuous domains such as bug populations, phytoremediation of metals, wound healing, maximization problems. The proof can be done by introducing restriction on the operator in several cases. The concepts in time scale version such as time scales calculus will be used that allows to unify and extend many problems from the theories of differential and of difference equations. In addition, using chain rule, and some properties of multiple integrals on time scales, some theorems of Fubini and the inequality of H¨older.

Keywords: time scales, inequality of hardy, inequality of coposon, steklov operator

Procedia PDF Downloads 53

Authors: Chih-Chun Tang

Abstract:

Nostalgia, to some people, may seem foolhardy in a way. However, nostalgia is a completely and intensely private but social, collective emotion. It has continuing consequence and outgrowth for our lives as social actions. It leads people to hunt and explore remembrance of persons and places of our past in an effort to confer meaning of persons and places of present. In the ‘Poems of 1912-13’ Thomas Hardy, a British poet, composed a series of poems after the unexpected death of his long-disaffected wife, Emma. The series interprets the cognitive and emotional concussion of Emma’s death on Hardy, concerning his mind and real visit to the landscape in Cornwall, England. Both spaces perform the author’s innermost in thought to his late wife and to the landscape. They present an apparent counterpart of the poet and his afflicted conscience. After Emma had died, Hardy carried her recollections alive by roaming about in the real visit and whimsical land (space) they once had drifted and meandered. This paper highlights the nostalgias and feds that seem endlessly to crop up.

Keywords: Thomas Hardy, remembrance, psychological, poems 1912-13, Fred Davis, nostalgia

Procedia PDF Downloads 236

Authors: Shams Alyusof

Abstract:

This work is to enrich the studies of the frames due to their prominent role in pure mathematics as well as in applied mathematics and many applications in computer science and engineering. Recently, there are remarkable studies of operators that preserve frames on some spaces, and this research could be considered as an extension of such studies. Indeed, this paper is to we characterize weighted composition operators that preserve frames in weighted Hardy spaces on the open unit disk. Moreover, it shows that this characterization does not apply to generalized weighted composition operators on such spaces. Nevertheless, this study could be extended to provide more specific characterizations.

Keywords: frames, generalized weighted composition operators, weighted Hardy spaces, analytic functions

Procedia PDF Downloads 80

Authors: Kairat T. Mynbaev, Elena N. Lomakina

Abstract:

We consider a Hardy-type operator generated by a family of open subsets of a Hausdorff topological space. The family is indexed with non-negative real numbers and is totally ordered. For this operator, we obtain two-sided bounds of its norm, a compactness criterion, and bounds for its approximation numbers. Previously, bounds for its approximation numbers have been established only in the one-dimensional case, while we do not impose any restrictions on the dimension of the Hausdorff space. The bounds for the norm and conditions for compactness earlier have been found using different methods by G. Sinnamon and K. Mynbaev. Our approach is different in that we use domain partitions for all problems under consideration.

Keywords: approximation numbers, boundedness and compactness, multidimensional Hardy operator, Hausdorff topological space

Procedia PDF Downloads 64

Authors: Mahdi Bazarganigilani

Abstract:

Aortic enlargement, also known as an aortic aneurysm, can occur when the walls of the aorta become weak. This disease can become deadly if overlooked and undiagnosed. In this paper, a computer-aided diagnosis (CAD) system was introduced to accurately diagnose aortic enlargement from chest x-ray images. An enhanced convolutional neural network (CNN) was employed and then trained by transfer learning by using three different main areas from the original images. The areas included the left lung, heart, and right lung. The accuracy of the system was then evaluated on 1001 samples by using 4-fold cross-validation. A promising accuracy of 90% was achieved in terms of the F-measure indicator. The results showed using different areas from the original image in the training phase of CNN could increase the accuracy of predictions. This encouraged the author to evaluate this method on a larger dataset and even on different CAD systems for further enhancement of this methodology.

Keywords: computer-aided diagnosis systems, aortic enlargement, chest X-ray, image processing, convolutional neural networks

Procedia PDF Downloads 121

Authors: Irina Peterburgsky

Abstract:

Let T be a unit circumference of a complex plane, E be a Banach space, E* and E** be its conjugate and second conjugate, respectively. In general, a Hardy space Hp(E), p ≥1, where functions act from the open unit disk to E, could contain a function for which even weak nontangential (angular) boundary value in the space E** does not exist at any point of the unit circumference T (C. Grossetete.) The situation is "better" when certain restrictions to the Banach space of values are applied (more or less resembling a classical case of scalar-valued functions depending on constrains, as shown by R. Ryan.) This paper shows that, nevertheless, in the case of a Banach space of a general type, the following positive statement is true: Proposition. For any function f(z) from Hp(E), p ≥ 1, there exists a function F(eiθ) on the unit circumference T to E** whose Poisson (in the Pettis sense) is integral regains the function f(z) on the open unit disk. Some characteristics of the function F(eiθ) are demonstrated.

Keywords: hardy spaces, Banach space-valued function, boundary values, Pettis integral

Procedia PDF Downloads 211

Authors: Leah Wilck

Abstract:

Eyewitness identification is arguably one of the most utilized practices within our legal system; however, exoneration cases indicate that this practice may lead to accuracy and conviction errors. The purpose of this study was to examine the effects of the cross-race effect, the phenomena in which people are able to more easily and accurately identify faces from within their racial category, on the accuracy of eyewitness identification. Participants watched three separate videos of a perpetrator trying to steal a bicycle. In each video, the perpetrator was of a different race and gender. Participants watched a video where the perpetrator was a Black male, a White male, and a White female. Following the completion of watching each video, participants were asked to recall everything they could about the perpetrator they witnessed. The initial results of the study did not find the expected cross-race effect impacted the eyewitness identification accuracy. These surprising results are discussed in terms of cross-race bias and recognition theory as well as applied implications.

Keywords: cross race effect, eyewitness identification, own-race bias, racial profiling

Procedia PDF Downloads 129

Authors: Konstantinos Perifanos, Eirini Florou, Dionysis Goutsos

Abstract:

This paper presents and benchmarks a number of end-to-end Deep Learning based models for metaphor detection in Greek. We combine Convolutional Neural Networks and Recurrent Neural Networks with representation learning to bear on the metaphor detection problem for the Greek language. The models presented achieve exceptional accuracy scores, significantly improving the previous state-of-the-art results, which had already achieved accuracy 0.82. Furthermore, no special preprocessing, feature engineering or linguistic knowledge is used in this work. The methods presented achieve accuracy of 0.92 and F-score 0.92 with Convolutional Neural Networks (CNNs) and bidirectional Long Short Term Memory networks (LSTMs). Comparable results of 0.91 accuracy and 0.91 F-score are also achieved with bidirectional Gated Recurrent Units (GRUs) and Convolutional Recurrent Neural Nets (CRNNs). The models are trained and evaluated only on the basis of training tuples, the related sentences and their labels. The outcome is a state-of-the-art collection of metaphor detection models, trained on limited labelled resources, which can be extended to other languages and similar tasks.

Keywords: metaphor detection, deep learning, representation learning, embeddings

Procedia PDF Downloads 109

Authors: Shiva Kumar, G. S. Vijay, Srinivas Pai P., Shrinivasa Rao B. R.

Abstract:

In the present study RBF neural networks were used for predicting the performance and emission parameters of a biodiesel engine. Engine experiments were carried out in a 4 stroke diesel engine using blends of diesel and Honge methyl ester as the fuel. Performance parameters like BTE, BSEC, Tech and emissions from the engine were measured. These experimental results were used for ANN modeling. RBF center initialization was done by random selection and by using Clustered techniques. Network was trained by using fixed and varying widths for the RBF units. It was observed that RBF results were having a good agreement with the experimental results. Networks trained by using clustering technique gave better results than using random selection of centers in terms of reduced MRE and increased prediction accuracy. The average MRE for the performance parameters was 3.25% with the prediction accuracy of 98% and for emissions it was 10.4% with a prediction accuracy of 80%.

Keywords: radial basis function networks, emissions, performance parameters, fuzzy c means

Procedia PDF Downloads 524

Authors: Levente Varga, Dávid Deritei, Mária Ercsey-Ravasz, Răzvan Florian, Zsolt I. Lázár, István Papp, Ferenc Járai-Szabó

Abstract:

One of the major shortcomings of widely used scientometric indicators is that different disciplines cannot be compared with each other. The issue of cross-disciplinary normalization has been long discussed, but even the classification of publications into scientific domains poses problems. Structural properties of citation networks offer new possibilities, however, the large size and constant growth of these networks asks for precaution. Here we present a new tool that in order to perform cross-field normalization of scientometric indicators of individual publications relays on the structural properties of citation networks. Due to the large size of the networks, a systematic procedure for identifying scientific domains based on a local community detection algorithm is proposed. The algorithm is tested with different benchmark and real-world networks. Then, by the use of this algorithm, the mechanism of the scientometric indicator normalization process is shown for a few indicators like the citation number, P-index and a local version of the PageRank indicator. The fat-tail trend of the article indicator distribution enables us to successfully perform the indicator normalization process.

Keywords: citation networks, cross-field normalization, local cluster detection, scientometric indicators

Procedia PDF Downloads 173

Authors: Zachary Pan

Abstract:

Our general mission is to solve the problem of classifying images into different body part types and deciding if each of them is healthy or not. However, for now, we will determine healthiness for only one-sixth of the body parts, specifically the chest. We will detect pneumonia in X-ray scans of those chest images. With this type of AI, doctors can use it as a second opinion when they are taking CT or X-ray scans of their patients. Another ad-vantage of using this machine learning classifier is that it has no human weaknesses like fatigue. The overall ap-proach to this problem is to split the problem into two parts: first, classify the image, then determine if it is healthy. In order to classify the image into a specific body part class, the body parts dataset must be split into test and training sets. We can then use many models, like neural networks or logistic regression models, and fit them using the training set. Now, using the test set, we can obtain a realistic accuracy the models will have on images in the real world since these testing images have never been seen by the models before. In order to increase this testing accuracy, we can also apply many complex algorithms to the models, like multiplicative weight update. For the second part of the problem, to determine if the body part is healthy, we can have another dataset consisting of healthy and non-healthy images of the specific body part and once again split that into the test and training sets. We then use another neural network to train on those training set images and use the testing set to figure out its accuracy. We will do this process only for the chest images. A major conclusion reached is that convolutional neural networks are the most reliable and accurate at image classification. In classifying the images, the logistic regression model, the neural network, neural networks with multiplicative weight update, neural networks with the black box algorithm, and the convolutional neural network achieved 96.83 percent accuracy, 97.33 percent accuracy, 97.83 percent accuracy, 96.67 percent accuracy, and 98.83 percent accuracy, respectively. On the other hand, the overall accuracy of the model that de-termines if the images are healthy or not is around 78.37 percent accuracy.

Keywords: body part, healthcare, machine learning, neural networks

Procedia PDF Downloads 68

Authors: Manoj Kumar

Abstract:

Ionization cross sections of molecules due to electron impact play an important role in chemical processes in various branches of applied physics, such as radiation chemistry, gas discharges, plasmas etching in semiconductors, planetary upper atmospheric physics, mass spectrometry, etc. In the present work, we have calculated the total ionization cross sections for Adenine (C₅H₅N₅), a biologically important molecule, by electron impact in the incident electron energy range from ionization threshold to 2 keV employing a well-known Jain-Khare semiempirical formulation based on Bethe and Möllor cross sections. In the non-availability of the experimental results, the present results are in good agreement qualitatively as well as quantitatively with available theoretical results. The present results drive our confidence for further investigation of complex bio-molecule with better accuracy. Notwithstanding, the present method can deduce reliable cross-sectional data for complex targets with adequate accuracy and may facilitate the acclimatization of calculated cross-sections into atomic molecular cross-section data sets for modeling codes and other applications.

Keywords: electron impact ionization cross-sections, oscillator strength, jain-khare semiempirical approach

Procedia PDF Downloads 78

Authors: Salah Ghamari

Abstract:

In this study, the ability of self organizing map artificial (SOM) neural networks in clustering the wheat seeds varieties according to morphological properties of them was considered. The SOM is one type of unsupervised competitive learning. Experimentally, five morphological features of 300 seeds (including three varieties: gaskozhen, Md and sardari) were obtained using image processing technique. The results show that the artificial neural network has a good performance (90.33% accuracy) in classification of the wheat varieties despite of high similarity in them. The highest classification accuracy (100%) was achieved for sardari.

Keywords: artificial neural networks, clustering, self organizing map, wheat variety

Procedia PDF Downloads 606

Authors: A. Sri Janani, K. Immanuel Arokia James

Abstract:

Cognitive radio ad hoc networks different unlicensed users may acquire different available channel sets. This non-uniform spectrum availability imposes special design challenges for broadcasting in CR ad hoc networks. Cognitive radio automatically detects available channels in wireless spectrum. This is a form of dynamic spectrum management. Cross-layer optimization is proposed, using this can allow far away secondary users can also involve into channel work. So it can increase the throughput and it will overcome the collision and time delay.

Keywords: cognitive radio, cross layer optimization, CR mesh network, heterogeneous spectrum, mesh topology, random routing optimization technique

Procedia PDF Downloads 331

Authors: Najmeh Mohsenifar, Narjes Mohsenifar, Abbas Kargar

Abstract:

In this paper, has been presented a stable method for predicting the ECG signals through the RBF neural networks, by the PSO algorithm. In spite of quasi-periodic ECG signal from a healthy person, there are distortions in electro cardiographic data for a patient. Therefore, there is no precise mathematical model for prediction. Here, we have exploited neural networks that are capable of complicated nonlinear mapping. Although the architecture and spread of RBF networks are usually selected through trial and error, the PSO algorithm has been used for choosing the best neural network. In this way, 2 second of a recorded ECG signal is employed to predict duration of 20 second in advance. Our simulations show that PSO algorithm can find the RBF neural network with minimum MSE and the accuracy of the predicted ECG signal is 97 %.

Keywords: electrocardiogram, RBF artificial neural network, PSO algorithm, predict, accuracy

Procedia PDF Downloads 587

Authors: Chad Goldsworthy, B. Rajeswari Matam

Abstract:

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

Keywords: convolutional neural networks, data imbalance, deep learning, focal loss, species classification, wildlife conservation

Procedia PDF Downloads 142

Authors: Nouria Bouchikhi, Soufiane Bedjaoui, C. Tewfik Bouchaour, Lamia Alachaher Bedjaoui, Ulrich Maschke

Abstract:

Swelling properties and phase diagrams of binary systems composed of liquid crystalline networks and a low molecular mass liquid crystal (LMWLC) have been investigated. The networks were prepared by ultraviolet (UV) irradiation of reactive mixtures including a monomer, a cross-linking agent and a photo-initiator. These networks were prepared using two cross-linking agents: 1,6 hexanedioldiacrylate (HDDA) and a mesogenic acrylic acid 6-(4’-(6-acryloyloxy-hexyloxy) biphenyl-4-yl oxy) hexyl ester (AHBH). The obtained dry networks were characterized by differential scanning calorimetry, and immersed in an excess of a LMWLC solvent 4-cyano-4’-pentylbiphenyl (5CB), forming polymer gels. A detailed study by polarized optical microscopy allowed to determine the swelling degree of the gels and to follow the phase behavior of the solvent inside the polymer matrix in a wide range of temperature. It has been found that the gels undergo a sharp decrease of their swelling degree in response to an infinitesimal change of temperature. This finding adds new and interesting aspects on the actuators applications. We have subsequently explored the effect of different parameters on volume phase transition of these liquid crystalline materials. Such as the cross-linking density (CD), a nature of cross-linking agent and the photo initiator concentration.

Keywords: cross-linking density, liquid crystalline elastomers, phase diagrams, swelling

Procedia PDF Downloads 294

Authors: Jonathan Sohn

Abstract:

Deep neural networks (DNNs) have shown state-of-the-art performance for many applications, including computer vision, natural language processing, and speech recognition. Recently, adversarial attacks have been studied in the context of deep neural networks, which aim to alter the results of deep neural networks by modifying the inputs slightly. For example, an adversarial attack on a DNN used for object detection can cause the DNN to miss certain objects. As a result, the reliability of DNNs is undermined by their lack of robustness against adversarial attacks, raising concerns about their use in safety-critical applications such as autonomous driving. In this paper, we focus on studying the adversarial attacks and defenses on DNNs for image classification. There are two types of adversarial attacks studied which are fast gradient sign method (FGSM) attack and projected gradient descent (PGD) attack. A DNN forms decision boundaries that separate the input images into different categories. The adversarial attack slightly alters the image to move over the decision boundary, causing the DNN to misclassify the image. FGSM attack obtains the gradient with respect to the image and updates the image once based on the gradients to cross the decision boundary. PGD attack, instead of taking one big step, repeatedly modifies the input image with multiple small steps. There is also another type of attack called the target attack. This adversarial attack is designed to make the machine classify an image to a class chosen by the attacker. We can defend against adversarial attacks by incorporating adversarial examples in training. Specifically, instead of training the neural network with clean examples, we can explicitly let the neural network learn from the adversarial examples. In our experiments, the digit recognition accuracy on the MNIST dataset drops from 97.81% to 39.50% and 34.01% when the DNN is attacked by FGSM and PGD attacks, respectively. If we utilize FGSM training as a defense method, the classification accuracy greatly improves from 39.50% to 92.31% for FGSM attacks and from 34.01% to 75.63% for PGD attacks. To further improve the classification accuracy under adversarial attacks, we can also use a stronger PGD training method. PGD training improves the accuracy by 2.7% under FGSM attacks and 18.4% under PGD attacks over FGSM training. It is worth mentioning that both FGSM and PGD training do not affect the accuracy of clean images. In summary, we find that PGD attacks can greatly degrade the performance of DNNs, and PGD training is a very effective way to defend against such attacks. PGD attacks and defence are overall significantly more effective than FGSM methods.

Keywords: deep neural network, adversarial attack, adversarial defense, adversarial machine learning

Procedia PDF Downloads 153

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: distillation, machine learning, neural networks, quantization

Procedia PDF Downloads 288

Authors: Kaoutar Abbahaddou, Mohammed Salah Chiadmi

Abstract:

The need to find a relevant way to predict the next-day price of a stock index is a real concern for many financial stakeholders and researchers. We have known across years the proliferation of several methods. Nevertheless, among all these methods, the most controversial one is a machine learning algorithm that claims to be reliable, namely neural networks. Thus, the purpose of this article is to study the prediction power of neural networks in the particular case of Islamic finance as it is an under-looked area. In this article, we will first briefly present a review of the literature regarding neural networks and Islamic finance. Next, we present the architecture and principles of artificial neural networks most commonly used in finance. Then, we will show its empirical application on two Islamic stock indexes. The accuracy rate would be used to measure the performance of the algorithm in predicting the right price the next day. As a result, we can conclude that artificial neural networks are a reliable method to predict the next-day price for Islamic indices as it is claimed for conventional ones.

Keywords: Islamic finance, stock price prediction, artificial neural networks, machine learning

Procedia PDF Downloads 191

Authors: Bakr Gomaa

Abstract:

Cross ventilation is an important and practical mean to achieve thermal comfort and conserve energy. This is especially true in the breezy waterfront settings. However, due to a number of factors, cross ventilation in buildings is usually studied by using oversimplified scenarios. It is then reasonable to study the impact of complex set of factors on the accuracy of predicting air flow rate because of wind driven cross ventilation. The objective of this paper is to provide architects with the tools necessary to achieve natural ventilation for cooling purposes in a waterfront urban canyon context. Also, urban canyons have not received much attention in terms of their impact on cross ventilation, and while we know how the wind flows between buildings in different urban canyon settings, the effect of the parallel-to-the-wind urban canyon on cross ventilation in buildings remains unclear. For this, we use detailed weather data, boundary layer correction factor, and CFD simulations to study the pressure patterns that form on the canyons surfaces in the case study of Alexandria. We found that the simplified numerical methods of calculating the cross ventilation in buildings can lead to inaccurate design decisions.

Keywords: cross ventilation, Alexandria, CFD, urban canyon

Procedia PDF Downloads 223

Authors: Jamerson Felipe Pereira Lima, Jeane Cecília Bezerra de Melo

Abstract:

The difficulty and cost related to obtaining of protein tertiary structure information through experimental methods, such as X-ray crystallography or NMR spectroscopy, helped raising the development of computational methods to do so. An approach used in these last is prediction of tridimensional structure based in the residue chain, however, this has been proved an NP-hard problem, due to the complexity of this process, explained by the Levinthal paradox. An alternative solution is the prediction of intermediary structures, such as the secondary structure of the protein. Artificial Intelligence methods, such as Bayesian statistics, artificial neural networks (ANN), support vector machines (SVM), among others, were used to predict protein secondary structure. Due to its good results, artificial neural networks have been used as a standard method to predict protein secondary structure. Recent published methods that use this technique, in general, achieved a Q3 accuracy between 75% and 83%, whereas the theoretical accuracy limit for protein prediction is 88%. Alternatively, to achieve better results, support vector machines prediction methods have been developed. The statistical evaluation of methods that use different AI techniques, such as ANNs and SVMs, for example, is not a trivial problem, since different training sets, validation techniques, as well as other variables can influence the behavior of a prediction method. In this study, we propose a prediction method based on artificial neural networks, which is then compared with a selected SVM method. The chosen SVM protein secondary structure prediction method is the one proposed by Huang in his work Extracting Physico chemical Features to Predict Protein Secondary Structure (2013). The developed ANN method has the same training and testing process that was used by Huang to validate his method, which comprises the use of the CB513 protein data set and three-fold cross-validation, so that the comparative analysis of the results can be made comparing directly the statistical results of each method.

Keywords: artificial neural networks, protein secondary structure, protein structure prediction, support vector machines

Procedia PDF Downloads 581

Authors: Ulas Vural, M. Ergun Okay, E. Mesut Yildiz

Abstract:

Telecommunication service providers demand accurate and precise prediction of customer churn probabilities to increase the effectiveness of their customer relation services. The large amount of customer data owned by the service providers is suitable for analysis by machine learning methods. In this study, expenditure data of customers are analyzed by using an artificial neural network (ANN). The ANN model is applied to the data of customers with different billing duration. The proposed model successfully predicts the churn probabilities at 83% accuracy for only three months expenditure data and the prediction accuracy increases up to 89% when the nine month data is used. The experiments also show that the accuracy of ANN model increases on an extended feature set with information of the changes on the bill amounts.

Keywords: customer relationship management, churn prediction, telecom industry, deep learning, artificial neural networks

Procedia PDF Downloads 116

Authors: K. A. Laptinskiy, S. A. Burikov, A. M. Vervald, S. A. Dolenko, T. A. Dolenko

Abstract:

The article presents the results of the application of artificial neural networks to separate the fluorescent contribution of nanodiamonds used as biomarkers, adsorbents and carriers of drugs in biomedicine, from a fluorescent background of own biological fluorophores. The principal possibility of solving this problem is shown. Use of neural network architecture let to detect fluorescence of nanodiamonds against the background autofluorescence of egg white with high accuracy - better than 3 ug/ml.

Keywords: artificial neural networks, fluorescence, data aggregation, biomarkers

Procedia PDF Downloads 668

Authors: Christina Adly, Meena Abdelmeseeh, Tamer Basha

Abstract:

This paper proposes a system for hand movement recognition using multichannel surface EMG(sEMG) signals obtained from 40 subjects using 40 different exercises, which are available on the Ninapro(Non-Invasive Adaptive Prosthetics) database. First, we applied processing methods to the raw sEMG signals to convert them to their amplitudes. Second, we used deep learning methods to solve our problem by passing the preprocessed signals to Fully connected neural networks(FCNN) and recurrent neural networks(RNN) with Long Short Term Memory(LSTM). Using intrasubject evaluation, The accuracy using the FCNN is 72%, with a processing time for training around 76 minutes, and for RNN's accuracy is 79.9%, with 8 minutes and 22 seconds processing time. Third, we applied some postprocessing methods to improve the accuracy, like majority voting(MV) and Movement Error Rate(MER). The accuracy after applying MV is 75% and 86% for FCNN and RNN, respectively. The MER value has an inverse relationship with the prediction delay while varying the window length for measuring the MV. The different part uses the RNN with the intersubject evaluation. The experimental results showed that to get a good accuracy for testing with reasonable processing time, we should use around 20 subjects.

Keywords: hand movement recognition, recurrent neural network, movement error rate, intrasubject evaluation, intersubject evaluation

Procedia PDF Downloads 102

Authors: Salah Hamri, Tewfik Bouchaour, Ulrich Maschke

Abstract:

The aim of this works is the study of swelling ratio of cross-linked polymer networks poly (2-hydroxyethyl methacrylate) (PHEMA). The system composed of erythrosine and Triethanolamine, in aqueous medium, is used as photo-initiator and 1,6-Hexanediol diacrylate as cross-linker. The analysis of UV-visible and infrared spectra, which were taken at different times during polymerization/cross linking, makes it possible to obtain useful information on the reaction mechanism. The swelling behavior was study by changing the nature of solvent, dye sensitizer (erythrosine, rose Bengal and eosin), and pH of the medium. The exploitation of experimental results using Fick diffusion model is also expected and shows a good correlation between theoretical and experimental results.

Keywords: cross-linker, photo-sensitizer, polymer network, swelling ratio

Procedia PDF Downloads 283

Authors: Nadia el Atlas, Mohammed el Aroussi, Mohammed Wahbi

Abstract:

In this paper a novel technique of mass characterization based on robust features-fusion is presented. The proposed method consists of mainly four stages: (a) the first phase involves segmenting the masses using edge information’s. (b) The second phase is to calculate and fuse the most relevant morphological features. (c) The last phase is the classification step which allows us to classify the images into benign and malignant masses. In this step we have implemented Support Vectors Machines (SVM) and Artificial Neural Networks (ANN), which were evaluated with the following performance criteria: confusion matrix, accuracy, sensitivity, specificity, receiver operating characteristic ROC, and error histogram. The effectiveness of this new approach was evaluated by a recently developed database: INBREAST database. The fusion of the most appropriate morphological features provided very good results. The SVM gives accuracy to within 64.3%. Whereas the ANN classifier gives better results with an accuracy of 97.5%.

Keywords: breast cancer, mammography, CAD system, features, fusion

Procedia PDF Downloads 564