Search results for: message passing neural network

Authors: Aisultan Shoiynbek, Darkhan Kuanyshbay, Paulo Menezes, Akbayan Bekarystankyzy, Assylbek Mukhametzhanov, Temirlan Shoiynbek

Abstract:

Speech emotion recognition (SER) has received increasing research interest in recent years. It is a common practice to utilize emotional speech collected under controlled conditions recorded by actors imitating and artificially producing emotions in front of a microphone. There are four issues related to that approach: emotions are not natural, meaning that machines are learning to recognize fake emotions; emotions are very limited in quantity and poor in variety of speaking; there is some language dependency in SER; consequently, each time researchers want to start work with SER, they need to find a good emotional database in their language. This paper proposes an approach to create an automatic tool for speech emotion extraction based on facial emotion recognition and describes the sequence of actions involved in the proposed approach. One of the first objectives in the sequence of actions is the speech detection issue. The paper provides a detailed description of the speech detection model based on a fully connected deep neural network for Kazakh and Russian. Despite the high results in speech detection for Kazakh and Russian, the described process is suitable for any language. To investigate the working capacity of the developed model, an analysis of speech detection and extraction from real tasks has been performed.

Keywords: deep neural networks, speech detection, speech emotion recognition, Mel-frequency cepstrum coefficients, collecting speech emotion corpus, collecting speech emotion dataset, Kazakh speech dataset

Procedia PDF Downloads 27

Authors: Zhaoxin Luo, Michael Zhu

Abstract:

In natural languages, there are always complex semantic hierarchies. Obtaining the feature representation based on these complex semantic hierarchies becomes the key to the success of the model. Several RNN models have recently been proposed to use latent indicators to obtain the hierarchical structure of documents. However, the model that only uses a single-layer latent indicator cannot achieve the true hierarchical structure of the language, especially a complex language like Chinese. In this paper, we propose a deep layered model that stacks arbitrarily many RNN layers equipped with latent indicators. After using EM and training it hierarchically, our model solves the computational problem of stacking RNN layers and makes it possible to stack arbitrarily many RNN layers. Our deep hierarchical model not only achieves comparable results to large pre-trained models on the Chinese short text classification problem but also achieves state of art results on the Chinese long text classification problem.

Keywords: nature language processing, recurrent neural network, hierarchical structure, document classification, Chinese

Procedia PDF Downloads 69

Authors: Saeed Alzahrani, Yaser Miaji

Abstract:

The project involves the design and simulation of a Mobile Cellular Telecommunication Network using the software tool CelPlanner. The design is mainly concerned with Global System for Mobile Communications . The design and simulation of the network is done for a small part of the area allocated for us in the terrain area of Shreveport city .The project is concerned with designing a network that is cost effective and which also efficiently meets the required Grade of Service (GOS) AND Quality of Service (QOS).The expected outcome of this project is the design of a network that gives a good coverage for the area allocated to us with minimum co-channel interference and adjacent channel interference. The Handover and Traffic Handling Capacity should also be taken into consideration and should be good for the given area . The Traffic Handling Capacity of the network in a way decides whether the designed network is good or bad . The design also takes into consideration the topographical and morphological information.

Keywords: mobile communication, GSM, radio base station, network planning

Procedia PDF Downloads 439

Authors: Mamouni Mohammed Dhiya Eddine

Abstract:

This work aims at studying the exploitation of high-speed networks of clusters for distributed storage. Parallel applications running on clusters require both high-performance communications between nodes and efficient access to the storage system. Many studies on network technologies led to the design of dedicated architectures for clusters with very fast communications between computing nodes. Efficient distributed storage in clusters has been essentially developed by adding parallelization mechanisms so that the server(s) may sustain an increased workload. In this work, we propose to improve the performance of distributed storage systems in clusters by efficiently using the underlying high-performance network to access distant storage systems. The main question we are addressing is: do high-speed networks of clusters fit the requirements of a transparent, efficient and high-performance access to remote storage? We show that storage requirements are very different from those of parallel computation. High-speed networks of clusters were designed to optimize communications between different nodes of a parallel application. We study their utilization in a very different context, storage in clusters, where client-server models are generally used to access remote storage (for instance NFS, PVFS or LUSTRE). Our experimental study based on the usage of the GM programming interface of MYRINET high-speed networks for distributed storage raised several interesting problems. Firstly, the specific memory utilization in the storage access system layers does not easily fit the traditional memory model of high-speed networks. Secondly, client-server models that are used for distributed storage have specific requirements on message control and event processing, which are not handled by existing interfaces. We propose different solutions to solve communication control problems at the filesystem level. We show that a modification of the network programming interface is required. Data transfer issues need an adaptation of the operating system. We detail several propositions for network programming interfaces which make their utilization easier in the context of distributed storage. The integration of a flexible processing of data transfer in the new programming interface MYRINET/MX is finally presented. Performance evaluations show that its usage in the context of both storage and other types of applications is easy and efficient.

Keywords: distributed storage, remote file access, cluster, high-speed network, MYRINET, zero-copy, memory registration, communication control, event notification, application programming interface

Procedia PDF Downloads 222

Authors: Varun Agarwal

Abstract:

Introduction: With the advent of personalized medicine, histopathological review of whole slide images (WSIs) for cancer diagnosis presents an exceedingly time-consuming, complex task. Specifically, detecting metastatic regions in WSIs of sentinel lymph node biopsies necessitates a full-scanned, holistic evaluation of the image. Thus, digital pathology, low-level image manipulation algorithms, and machine learning provide significant advancements in improving the efficiency and accuracy of WSI analysis. Using Camelyon16 data, this paper proposes a deep learning pipeline to automate and ameliorate breast cancer metastasis localization and WSI classification. Methodology: The model broadly follows five stages -region of interest detection, WSI partitioning into image tiles, convolutional neural network (CNN) image-segment classifications, probabilistic mapping of tumor localizations, and further processing for whole WSI classification. Transfer learning is applied to the task, with the implementation of Inception-ResNetV2 - an effective CNN classifier that uses residual connections to enhance feature representation, adding convolved outputs in the inception unit to the proceeding input data. Moreover, in order to augment the performance of the transfer learning CNN, a stack of convolutional denoising autoencoders (CDAE) is applied to produce embeddings that enrich image representation. Through a saliency-detection algorithm, visual training segments are generated, which are then processed through a denoising autoencoder -primarily consisting of convolutional, leaky rectified linear unit, and batch normalization layers- and subsequently a contrast-normalization function. A spatial pyramid pooling algorithm extracts the key features from the processed image, creating a viable feature map for the CNN that minimizes spatial resolution and noise. Results and Conclusion: The simplified and effective architecture of the fine-tuned transfer learning Inception-ResNetV2 network enhanced with the CDAE stack yields state of the art performance in WSI classification and tumor localization, achieving AUC scores of 0.947 and 0.753, respectively. The convolutional feature retention and compilation with the residual connections to inception units synergized with the input denoising algorithm enable the pipeline to serve as an effective, efficient tool in the histopathological review of WSIs.

Keywords: breast cancer, convolutional neural networks, metastasis mapping, whole slide images

Procedia PDF Downloads 131

Authors: Maryam Zarabian, Hector Guzman, Pedro Pereira-Almao, Abraham Fapojuwo

Abstract:

Dry reforming of methane (DRM) has sparked significant industrial and scientific interest not only as a viable alternative for addressing the environmental concerns of two main contributors of the greenhouse effect, i.e., carbon dioxide (CO₂) and methane (CH₄), but also produces syngas, i.e., a mixture of hydrogen (H₂) and carbon monoxide (CO) utilized by a wide range of downstream processes as a feedstock for other chemical productions. In this study, we develop an AI-enable syngas production model to tackle the problem of achieving an equivalent H₂/CO ratio [1:1] with respect to the most efficient conversion. Firstly, the unsupervised density-based spatial clustering of applications with noise (DBSAN) algorithm removes outlier data points from the original experimental dataset. Then, random forest (RF) and deep neural network (DNN) models employ the error-free dataset to predict the DRM results. DNN models inherently would not be able to obtain accurate predictions without a huge dataset. To cope with this limitation, we employ reusing pre-trained layers’ approaches such as transfer learning and greedy layer-wise pretraining. Compared to the other deep models (i.e., pure deep model and transferred deep model), the greedy layer-wise pre-trained deep model provides the most accurate prediction as well as similar accuracy to the RF model with R² values 1.00, 0.999, 0.999, 0.999, 0.999, and 0.999 for the total outlet flow, H₂/CO ratio, H₂ yield, CO yield, CH₄ conversion, and CO₂ conversion outputs, respectively.

Keywords: artificial intelligence, dry reforming of methane, artificial neural network, deep learning, machine learning, transfer learning, greedy layer-wise pretraining

Procedia PDF Downloads 88

Authors: Dulal Bhaumik, Fei Jie, Runa Bhaumik, Bikas Sinha

Abstract:

Human brain is an amazingly complex network. Aberrant activities in this network can lead to various neurological disorders such as multiple sclerosis, Parkinson’s disease, Alzheimer’s disease and autism. fMRI has emerged as an important tool to delineate the neural networks affected by such diseases, particularly autism. In this paper, we propose mixed-effects models together with an appropriate procedure for controlling false discoveries to detect disrupted connectivities in whole brain studies. Results are illustrated with a large data set known as Autism Brain Imaging Data Exchange or ABIDE which includes 361 subjects from 8 medical centers. We believe that our findings have addressed adequately the small sample inference problem, and thus are more reliable for therapeutic target for intervention. In addition, our result can be used for early detection of subjects who are at high risk of developing neurological disorders.

Keywords: ABIDE, autism spectrum disorder, fMRI, mixed-effects model

Procedia PDF Downloads 291

Authors: Bubai Maji, Monorama Swain

Abstract:

Speech is considered an essential and most natural medium for the interaction between machines and humans. However, extracting effective features for speech emotion recognition (SER) is remains challenging. The present studies show that the temporal information captured but high-level temporal-feature learning is yet to be investigated. In this paper, we present an efficient novel method using the Self-attention (SA) mechanism in a combination of Convolutional Neural Network (CNN) and Bi-directional Gated Recurrent Unit (Bi-GRU) network to learn high-level temporal-feature. In order to further enhance the representation of the high-level temporal-feature, we integrate a Bi-GRU output with learnable weights features by SA, and improve the performance. We evaluate our proposed method on our created SITB-OSED and IEMOCAP databases. We report that the experimental results of our proposed method achieve state-of-the-art performance on both databases.

Keywords: Bi-GRU, 1D-CNNs, self-attention, speech emotion recognition

Procedia PDF Downloads 114

Authors: F. Ghazalnaz Sharifonnasabi, Iman Makhdoom

Abstract:

Breast cancer is a serious health concern that affects many people around the world. According to a study published in the Breast journal, the global burden of breast cancer is expected to increase significantly over the next few decades. The number of deaths from breast cancer has been increasing over the years, but the age-standardized mortality rate has decreased in some countries. It’s important to be aware of the risk factors for breast cancer and to get regular check- ups to catch it early if it does occur. Machin learning techniques have been used to aid in the early detection and diagnosis of breast cancer. These techniques, that have been shown to be effective in predicting and diagnosing the disease, have become a research hotspot. In this study, we consider two deep learning approaches including: Multi-Layer Perceptron (MLP), and Convolutional Neural Network (CNN). We also considered the five-machine learning algorithm titled: Decision Tree (C4.5), Naïve Bayesian (NB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) Algorithm and XGBoost (eXtreme Gradient Boosting) on the Breast Cancer Wisconsin Diagnostic dataset. We have carried out the process of evaluating and comparing classifiers involving selecting appropriate metrics to evaluate classifier performance and selecting an appropriate tool to quantify this performance. The main purpose of the study is predicting and diagnosis breast cancer, applying the mentioned algorithms and also discovering of the most effective with respect to confusion matrix, accuracy and precision. It is realized that CNN outperformed all other classifiers and achieved the highest accuracy (0.982456). The work is implemented in the Anaconda environment based on Python programing language.

Keywords: breast cancer, multi-layer perceptron, Naïve Bayesian, SVM, decision tree, convolutional neural network, XGBoost, KNN

Procedia PDF Downloads 78

Authors: Kenichiro Hida, Shin-Ichi Kuribayashi

Abstract:

In a conventional network, most network devices, such as routers, are dedicated devices that do not have much variation in capacity. In recent years, a new concept of network functions virtualisation (NFV) has come into use. The intention is to implement a variety of network functions with software on general-purpose servers and this allows the network operator to select their capacities and locations without any constraints. This paper focuses on the allocation of NFV-based routing functions which are one of critical network functions, and presents the virtual routing function allocation algorithm that minimizes the total power consumption. In addition, this study presents the useful allocation policy of virtual routing functions, based on an evaluation with a ladder-shaped network model. This policy takes the ratio of the power consumption of a routing function to that of a circuit and traffic distribution between areas into consideration. Furthermore, the present paper shows that there are cases where the use of NFV-based routing functions makes it possible to reduce the total power consumption dramatically, in comparison to a conventional network, in which it is not economically viable to distribute small-capacity routing functions.

Keywords: NFV, resource allocation, virtual routing function, minimum power consumption

Procedia PDF Downloads 344

Authors: Afroze Ibrahim Baqapuri, Ilya Trofimov

Abstract:

Sponsored search is a multi-billion dollar industry and makes up a major source of revenue for search engines (SE). Click-through-rate (CTR) estimation plays a crucial role for ads selection, and greatly affects the SE revenue, advertiser traffic and user experience. We propose a novel architecture of solving CTR prediction problem by combining artificial neural networks (ANN) with decision trees. First, we compare ANN with respect to other popular machine learning models being used for this task. Then we go on to combine ANN with MatrixNet (proprietary implementation of boosted trees) and evaluate the performance of the system as a whole. The results show that our approach provides a significant improvement over existing models.

Keywords: neural networks, sponsored search, web advertisement, click prediction, click-through rate

Procedia PDF Downloads 574

Authors: Abe D. Desta

Abstract:

This study suggests a deep learning-based method for identifying and following moving objects in surveillance video. The proposed method uses a fast regional convolution neural network (F-RCNN) trained on a substantial dataset of vehicle images to first detect vehicles. A Kalman filter and a data association technique based on a Hungarian algorithm are then used to monitor the observed vehicles throughout time. However, in general, F-RCNN algorithms have been shown to be effective in achieving high detection accuracy and robustness in this research study. For example, in one study The study has shown that the vehicle detection and tracking, the system was able to achieve an accuracy of 97.4%. In this study, the F-RCNN algorithm was compared to other popular object detection algorithms and was found to outperform them in terms of both detection accuracy and speed. The presented system, which has application potential in actual surveillance systems, shows the usefulness of deep learning approaches in vehicle detection and tracking.

Keywords: artificial intelligence, computer vision, deep learning, fast-regional convolutional neural networks, feature extraction, vehicle tracking

Procedia PDF Downloads 129

Authors: Zdravko Eškinja, Lovre Miljanić, Ognjen Kuljača

Abstract:

This paper is an overview of simulation tools used to model specific thermal dynamics that occurs while controlling double skin façade. Research has been conducted on simplified construction with single zone where one side is glazed. Heat flow and temperature responses are simulated in three different simulation tools: IDA-ICE, EnergyPlus and HAMBASE. The excitation of observed system, used in all simulations, was a temperature step of exterior environment. Air infiltration, insulation and other disturbances are excluded from this research. Although such isolated behaviour is not possible in reality, experiments are carried out to gain novel information about heat flow transients which are not observable under regular conditions. Results revealed new possibilities for adapting the parameters of the neural network regulator. Along numerical simulations, the same set-up has been also tested in a real-time experiment with a 1:18 scaled model and thermal chamber. The comparison analysis brings out interesting conclusion about simulation accuracy in this particular case.

Keywords: double skin façade, experimental tests, heat control, heat flow, simulated tests, simulation tools

Procedia PDF Downloads 233

Authors: Urvi Panwar, Kanchan Mishra, Kanjaksha Ghosh, ShankerLal Kothari

Abstract:

Background: Day-by-day the increasing prevalence of neurodegenerative diseases have become a global issue to manage them by medical sciences. The adult neural stem cells are rare and require an invasive and painful procedure to obtain it from central nervous system. Mesenchymal stem cell (MSCs) therapies have shown remarkable application in treatment of various cell injuries and cell loss. MSCs can be derived from various sources like adult tissues, human bone marrow, umbilical cord blood and cord tissue. MSCs have similar proliferation and differentiation capability, but the human umbilical cord-derived mesenchymal stem cells (hUCMSCs) are proved to be more beneficial with respect to cell procurement, differentiation to other cells, preservation, and transplantation. Material and method: Human umbilical cord is easily obtainable and non-controversial comparative to bone marrow and other adult tissues. The umbilical cord can be collected after delivery of baby, and its tissue can be cultured using explant culture method. Cell culture medium such as DMEMF12+10% FBS and DMEMF12+Neural growth factors (bFGF, human noggin, B27) with antibiotics (Streptomycin/Gentamycin) were used to culture and differentiate mesenchymal stem cells into neural cells, respectively. The characterisations of MSCs were done with Flow Cytometer for surface markers CD90, CD73 and CD105 and colony forming unit assay. The differentiated various neural cells will be characterised by fluorescence markers for neurons, astrocytes, and oligodendrocytes; quantitative PCR for genes Nestin and NeuroD1 and Western blotting technique for gap43 protein. Result and discussion: The high quality and number of MSCs were isolated from human umbilical cord via explant culture method. The obtained MSCs were differentiated into neural cells like neurons, astrocytes and oligodendrocytes. The differentiated neural cells can be used to treat neural injuries and neural cell loss by delivering cells by non-invasive administration via cerebrospinal fluid (CSF) or blood. Moreover, the MSCs can also be directly delivered to different injured sites where they differentiate into neural cells. Therefore, human umbilical cord is demonstrated to be an inexpensive and easily available source for MSCs. Moreover, the hUCMSCs can be a potential source for neural cell therapies and neural cell regeneration for neural cell injuries and neural cell loss. This new way of research will be helpful to treat and manage neural cell damages and neurodegenerative diseases like Alzheimer and Parkinson. Still the study has a long way to go but it is a promising approach for many neural disorders for which at present no satisfactory management is available.

Keywords: bone marrow, cell therapy, explant culture method, flow cytometer, human umbilical cord, mesenchymal stem cells, neurodegenerative diseases, neuroprotective, regeneration

Procedia PDF Downloads 202

Authors: S. Mousavian, A. Abedianpour, A. Khanmohammadi, S. Hematian, Gh. Eidi Veisi

Abstract:

Ionic liquids are finding a wide range of applications from reaction media to separations and materials processing. In these applications, Vapor–Liquid equilibrium (VLE) is the most important one. VLE for six systems at 353 K and activity coefficients at infinite dilution 〖(γ〗_i^∞) for various solutes (alkanes, alkenes, cycloalkanes, cycloalkenes, aromatics, alcohols, ketones, esters, ethers, and water) in the ionic liquids (1-ethyl-3-methylimidazolium bis (trifluoromethylsulfonyl)imide [EMIM][BTI], 1-hexyl-3-methyl imidazolium bis (trifluoromethylsulfonyl) imide [HMIM][BTI], 1-octyl-3-methylimidazolium bis(trifluoromethylsulfonyl) imide [OMIM][BTI], and 1-butyl-1-methylpyrrolidinium bis (trifluoromethylsulfonyl) imide [BMPYR][BTI]) have been used to train neural networks in the temperature range from (303 to 333) K. Densities of the ionic liquids, Hildebrant constant of substances, and temperature were selected as input of neural networks. The networks with different hidden layers were examined. Networks with seven neurons in one hidden layer have minimum error and good agreement with experimental data.

Keywords: ionic liquid, neural networks, VLE, dilute solution

Procedia PDF Downloads 301

Authors: Hanbey Hazar, Hakan Gul

Abstract:

In this study, Thermal Barrier Coating (TBC) application is performed in order to reduce the engine emissions. Piston, exhaust, and intake valves of a single-cylinder four-cycle gasoline engine were coated with chromium carbide (Cr3C2) at a thickness of 300 µm by using the Plasma Spray coating method which is a TBC method. Gasoline engine was converted into an LPG system. The study was conducted in 4 stages. In the first stage, the piston, exhaust, and intake valves of the gasoline engine were coated with Cr3C2. In the second stage, gasoline engine was converted into the LPG system and the emission values in this engine were recorded. In the third stage, the experiments were repeated under the same conditions with a standard (uncoated) engine and the results were recorded. In the fourth stage, data obtained from both engines were loaded on Artificial Neural Networks (ANN) and estimated values were produced for every revolution. Thus, mathematical modeling of coated and uncoated engines was performed by using ANN. While there was a slight increase in exhaust gas temperature (EGT) of LPG engine due to TBC, carbon monoxide (CO) values decreased.

Keywords: LPG fuel, thermal barrier coating, artificial neural network, mathematical modelling

Procedia PDF Downloads 427

Authors: Bjørn Salskov, Nicolaj Bang, Charlotte Falko

Abstract:

Future labour markets demand employees that can carry out a non-linear task which is still not possible for computers. This means that employees must have well-developed soft-skills to perform at high levels in such a work environment. One of these soft-skills is presenting a message effectively. To be able to present a message effectively, one needs to practice this. To practice effectively, the trainee needs feedback on the current performance. Here VR environments can be used as a practice tool because it gives the trainee a sense of presence and reality. VR environments are becoming a cost-effective training method since it does not demand the presence of an expert to provide this feedback. The research article analysed in this study suggests that VR environment can be used and are able to provide the necessary feedback to the trainee which in turn will help the trainee become better at the task. The research analysed in this review does, however, show that there is a need for a study with larger sample size and a study which runs over a longer period.

Keywords: training, presentation, presentation skills, VR training, VR as a training tool, VR and presentation

Procedia PDF Downloads 122

Authors: Vishnu Pratap Singh Kirar

Abstract:

Wireless sensor network (WSN) is a network of many interconnected networked systems, they equipped with energy resources and they are used to detect other physical characteristics. On WSN, there are many researches are performed in past decades. WSN applicable in many security systems govern by military and in many civilian related applications. Thus, the security of WSN gets attention of researchers and gives an opportunity for many future aspects. Still, there are many other issues are related to deployment and overall coverage, scalability, size, energy efficiency, quality of service (QoS), computational power and many more. In this paper we discus about various applications and security related issue and requirements of WSN.

Keywords: wireless sensor network (WSN), wireless network attacks, wireless network security, security requirements

Procedia PDF Downloads 492

Authors: Syed Asif Hassan, Syed Atif Hassan

Abstract:

Multidrug-resistant Tuberculosis (MDR-TB) is an infection caused by the resistant strains of Mycobacterium tuberculosis that do not respond either to isoniazid or rifampicin, which are the most important anti-TB drugs. The increase in the occurrence of a drug-resistance strain of MTB calls for an intensive search of novel target-based therapeutics. In this context LprG (Rv1411c) a lipoprotein from MTB plays a pivotal role in the immune evasion of Mtb leading to survival and propagation of the bacterium within the host cell. Therefore, a machine learning method will be developed for generating a computational model that could predict for a potential anti LprG activity of the novel antituberculosis compound. The present study will utilize dataset from PubChem database maintained by National Center for Biotechnology Information (NCBI). The dataset involves compounds screened against MTB were categorized as active and inactive based upon PubChem activity score. PowerMV, a molecular descriptor generator, and visualization tool will be used to generate the 2D molecular descriptors for the actives and inactive compounds present in the dataset. The 2D molecular descriptors generated from PowerMV will be used as features. We feed these features into three different classifiers, namely, random forest, a deep neural network, and a recurring neural network, to build separate predictive models and choosing the best performing model based on the accuracy of predicting novel antituberculosis compound with an anti LprG activity. Additionally, the efficacy of predicted active compounds will be screened using SMARTS filter to choose molecule with drug-like features.

Keywords: antituberculosis drug, classifier, machine learning, molecular descriptors, prediction

Procedia PDF Downloads 392

Authors: Riad

Abstract:

As present world needs more and more energy in a low cost way, it needs to find out the optimal way of power generation. In the sense of low cost, renewable energy is one of the greatest sources of power generation. Water vapour of sea/river/lake can be used for power generation by using the greenhouse effect in a large flat type water chamber floating on the water surface. The water chamber will always be kept half filled. When water evaporates by sunlight, the high pressured gaseous water will be stored in the chamber. By passing through a pipe and by using aerodynamics it can be used for power generation. The water level of the chamber is controlled by some means. As a large amount of water evaporates, an estimation can be highlighted, approximately 3 to 4 thousand gallons of water evaporates from per acre of surface (this amount will be more by greenhouse effect). This large amount of gaseous water can be utilized for power generation by passing through a pipe. This method can be a source of power generation.

Keywords: renewable energy, greenhouse effect, water chamber, water vapour

Procedia PDF Downloads 356

Authors: Christoph Linse, Thomas Martinetz

Abstract:

Recent findings have shown that Neural Networks generalize also in over-parametrized regimes with zero training error. This is surprising, since it is completely against traditional machine learning wisdom. In our empirical study we fortify these findings in the domain of fine-grained image classification. We show that very large Convolutional Neural Networks with millions of weights do learn with only a handful of training samples and without image augmentation, explicit regularization or pretraining. We train the architectures ResNet018, ResNet101 and VGG19 on subsets of the difficult benchmark datasets Caltech101, CUB_200_2011, FGVCAircraft, Flowers102 and StanfordCars with 100 classes and more, perform a comprehensive comparative study and draw implications for the practical application of CNNs. Finally, we show that VGG19 with 140 million weights learns to distinguish airplanes and motorbikes with up to 95% accuracy using only 20 training samples per class.

Keywords: convolutional neural networks, fine-grained image classification, generalization, image recognition, over-parameterized, small data sets

Procedia PDF Downloads 90

Authors: S. Chebbi, A. Meddeb

Abstract:

The distribution networks are often exposed to harmful incidents which can halt the electricity supply of the customer. In this context, we studied a real case of a critical zone of the Tunisian network which is currently characterized by the dysfunction of its plan of protection. In this paper, we were interested in the harmonization of the protection plan settings in order to ensure a perfect selectivity and a better continuity of service on the whole of the network.

Keywords: distribution network Gabes-Tunisia, continuity of service, protection plan settings, selectivity

Procedia PDF Downloads 511

Authors: Ahmad Lutfi, Nikolas Dhega

Abstract:

The Malala deposit in northwest Sulawesi is the only known porphyry molybdenum and the only source for rhenium, occurrence in Indonesia. The neural network method produces results that correspond very closely to those of the knowledge-based fuzzy logic method and weights of evidence method. This method required data of solid geology, regional faults, airborne magnetic, gamma-ray survey data and GIS data. This interpretation of the network output fits with the intuitive notion that a prospective area has characteristics that closely resemble areas known to contain mineral deposits. Contrasts with the weights of evidence and fuzzy logic methods, where, for a given grid location, each input-parameter value automatically results in an increase in the prospective estimated. Malala District indicated molybdenum anomalies in stream sediments from in excess of 15 km2 were obtained, including the Takudan Fault as most prominent structure with striking 40̊ to 60̊ over a distance of about 30 km and in most places weakly at anomaly B, developed over an area of 4 km2, with a ‘shell’ up to 50 m thick at the intrusive contact with minor mineralization occurring in the Tinombo Formation. Series of NW trending, steeply dipping fracture zones, named the East Zone has an estimated resource of 100 Mt at 0.14% MoS2 and minimum target of 150 Mt 0.25%. The Malala porphyries occur as stocks and dykes with predominantly granitic, with fluorine-poor class of molybdenum deposits and belongs to the plutonic sub-type. Unidirectional solidification textures consisting of subparallel, crenulated layers of quartz that area separated by layers of intrusive material textures. The deuteric nature of the molybdenum mineralization and the dominance of carbonate alteration.The nature of the Stage I with alteration barren quartz K‐feldspar; and Stage II with alteration quartz‐K‐feldspar‐apatite-molybdenite veins combined with the presence of disseminated molybdenite with primary biotite in the host intrusive.

Keywords: molybdenite, Malala, porphyries, anomaly B

Procedia PDF Downloads 153

Authors: Farid Khosravikia, Patricia Clayton

Abstract:

The main objective of this work is to evaluate the advantages and disadvantages of various machine learning techniques in two key steps of seismic hazard and risk assessment of different types of structures. The first step is the development of ground-motion models, which are used for forecasting ground-motion intensity measures (IM) given source characteristics, source-to-site distance, and local site condition for future events. IMs such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available. Second, it is investigated how machine learning techniques could be beneficial for developing probabilistic seismic demand models (PSDMs), which provide the relationship between the structural demand responses (e.g., component deformations, accelerations, internal forces, etc.) and the ground motion IMs. In the risk framework, such models are used to develop fragility curves estimating exceeding probability of damage for pre-defined limit states, and therefore, control the reliability of the predictions in the risk assessment. In this study, machine learning algorithms like artificial neural network, random forest, and support vector machine are adopted and trained on the demand parameters to derive PSDMs for them. It is observed that such models can provide more accurate estimates of prediction in relatively shorter about of time compared to conventional methods. Moreover, they can be used for sensitivity analysis of fragility curves with respect to many modeling parameters without necessarily requiring more intense numerical response-history analysis.

Keywords: artificial neural network, machine learning, random forest, seismic risk analysis, seismic hazard analysis, support vector machine

Procedia PDF Downloads 106

Authors: Tian Lan, Yixiang Wang, Wenxin Tai, Yilan Lyu, Zufeng Wu

Abstract:

The monaural speech enhancement task in the time-frequencydomain has a myriad of approaches, with the stacked con-volutional neural network (CNN) demonstrating superiorability in feature extraction and selection. However, usingstacked single convolutions method limits feature represen-tation capability and generalization ability. In order to solvethe aforementioned problem, we propose an attention-basedadaptive convolutional network that integrates the multi-scale convolutional operations into a operation-specific blockvia input dependent attention to adapt to complex auditoryscenes. In addition, we introduce a two-stage progressivelearning method to enlarge the receptive field without a dra-matic increase in computation burden. We conduct a series ofexperiments based on the TIMIT corpus, and the experimen-tal results prove that our proposed model is better than thestate-of-art models on all metrics.

Keywords: speech enhancement, adaptive convolu-tion, progressive learning, time-frequency domain

Procedia PDF Downloads 124

Authors: John M. Lyne, K. Andrea Scott

Abstract:

In fluid dynamics, direct numerical simulations (DNS) of turbulent flows require large amounts of nodes to appropriately resolve all scales of energy transfer. Due to the size of these databases, sharing these datasets amongst the academic community is a challenge. Recent work has been done to investigate the use of super-resolution to enable database sharing, where a low-resolution flow field is super-resolved to high resolutions using a neural network. Recently, Generative Adversarial Networks (GAN) have grown in popularity with impressive results in the generation of faces, landscapes, and more. This work investigates the generation of unique high-resolution channel flow velocity fields from a low-dimensional latent space using a GAN. The training objective of the GAN is to generate samples in which the distribution of the generated samplesis ideally indistinguishable from the distribution of the training data. In this study, the network is trained using samples drawn from a statistically stationary channel flow at a Reynolds number of 560. Results show that the turbulent statistics and energy spectra of the generated flow fields are within reasonable agreement with those of the DNS data, demonstrating that GANscan produce the intricate multi-scale phenomena of turbulence.

Keywords: computational fluid dynamics, channel flow, turbulence, generative adversarial network

Procedia PDF Downloads 207

Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan

Abstract:

The most aggressive form of brain tumor is called glioma. Glioma is kind of tumor that arises from glial tissue of the brain and occurs quite often. A fully automatic 2D-CNN model for brain tumor segmentation is presented in this paper. We performed pre-processing steps to remove noise and intensity variances using N4ITK and standard intensity correction, respectively. We used Keras open-source library with Theano as backend for fast implementation of CNN model. In addition, we used BRATS 2015 MRI dataset to evaluate our proposed model. Furthermore, we have used SimpleITK open-source library in our proposed model to analyze images. Moreover, we have extracted random 2D patches for proposed 2D-CNN model for efficient brain segmentation. Extracting 2D patched instead of 3D due to less dimensional information present in 2D which helps us in reducing computational time. Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.77 for complete, 0.76 for core, 0.77 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.

Keywords: brain tumor segmentation, convolutional neural networks, deep learning, LGG

Procedia PDF Downloads 183

Authors: Juha Kettunen

Abstract:

The study analyzes the quality and the size of the strategic network of higher education institutions and the concept of fitness for purpose in quality assurance. It also analyses the transaction costs of networking that have consequences on the number of members in the network. Empirical evidence is presented from the Consortium on Applied Research and Professional Education, which is a European strategic network of six higher education institutions. The results of the study support the argument that the number of members in the strategic network should be relatively small to provide high-quality results. The practical importance is that networking has been able to promote international research and development projects. The results of this study are important for those who want to design and improve international networks in higher education.

Keywords: higher education, network, research and development, strategic management

Procedia PDF Downloads 350

Authors: Wafa’ Alsharafat, Khalid Batiha, Alaa Kassab

Abstract:

Wireless sensor network (WSN) is comprised of a huge number of small and cheap devices known as sensor nodes. Usually, these sensor nodes are massively and deployed randomly as in Ad-hoc over hostile and harsh environment to sense, collect and transmit data to the needed locations (i.e., base station). One of the main advantages of WSN is that the ability to work in unattended and scattered environments regardless the presence of humans such as remote active volcanoes environments or earthquakes. In WSN expanding network, lifetime is a major concern. Clustering technique is more important to maximize network lifetime. Nature-inspired algorithms are developed and optimized to find optimized solutions for various optimization problems. We proposed Energy Efficient Firefly Algorithm to improve network lifetime as long as possible.

Keywords: wireless network, SN, Firefly, energy efficiency

Procedia PDF Downloads 389

Authors: Gabriele Borg, Alexei Debono, Charlie Abela

Abstract:

There on a constant rise in the availability of high volumes of data gathered from multiple sources, resulting in an abundance of unprocessed information that can be used to monitor patterns and trends in user behaviour. Similarly, year after year, Malta is also constantly experiencing ongoing population growth and an increase in mobilization demand. This research takes advantage of data which is continuously being sourced and converting it into useful information related to the traffic problem on the Maltese roads. The scope of this paper is to provide a methodology to create a custom dataset (MalTra - Malta Traffic) compiled from multiple participants from various locations across the island to identify the most common routes taken to expose the main areas of activity. This use of big data is seen being used in various technologies and is referred to as ITSs (Intelligent Transportation Systems), which has been concluded that there is significant potential in utilising such sources of data on a nationwide scale. Furthermore, a series of traffic prediction graph neural network models are conducted to compare MalTra to large-scale traffic datasets.

Keywords: graph neural networks, traffic management, big data, mobile data patterns

Procedia PDF Downloads 133