A three-stage learning algorithm for deep multilayer perceptron (DMLP) with effective weight initialisation based on sparse auto-encoder is proposed in this paper, which aims to overcome difficulties in training deep neural networks with limited training data in high-dimensional feature space. At the first stage, unsupervised learning is adopted using sparse auto-encoder to obtain the initial weights of the feature extraction layers of the DMLP. At the second stage, error back-propagation is used to train the DMLP by fixing the weights obtained at the first stage for its feature extraction layers. At the third stage, all the weights of the DMLP obtained at the second stage are refined by error back-propagation. Network structures an

   The Compressional-wave (Vp) data are useful for reservoir exploration, drilling operations, stimulation, hydraulic fracturing employment, and development plans for a specific reservoir. Due to the different nature and behavior of the influencing parameters, more complex nonlinearity exists for Vp modeling purposes. In this study, a statistical relationship between compressional wave velocity and petrophysical parameters was developed from wireline log data for Jeribe formation in Fauqi oil field south Est Iraq, which is studied using single and multiple linear regressions. The model concentrated on predicting compressional wave velocity from petrophysical parameters and any pair of shear waves velocity, porosity, density, and

   The Compressional-wave (Vp) data are useful for reservoir exploration, drilling operations, stimulation, hydraulic fracturing employment, and development plans for a specific reservoir. Due to the different nature and behavior of the influencing parameters, more complex nonlinearity exists for Vp modeling purposes. In this study, a statistical relationship between compressional wave velocity and petrophysical parameters was developed from wireline log data for Jeribe formation in Fauqi oil field south Est Iraq, which is studied using single and multiple linear regressions. The model concentrated on predicting compressional wave velocity from petrophysical parameters and any pair of shear waves velocity, porosity, density, a

In this work, a single pile is physically modeled and embedded in an upper liquefiable loose sand layer overlying a non-liquefiable dense layer. A laminar soil container is adopted to simulate the coupled static-dynamic loading pile response during earthquake motions: Ali Algharbi, Halabjah, El-Centro, and Kobe earthquakes. During seismic events with combined loading, the rotation along the pile, the lateral and vertical displacements at the pile head as well as the pore pressure ratio in loose sandy soil were assessed. According to the experimental findings, combined loading that ranged from 50 to 100% of axial load would alter the pile reaction by reducing the pile head peak ground acceleration, rotation of the pile, and lateral displacem

Imitation learning is an effective method for training an autonomous agent to accomplish a task by imitating expert behaviors in their demonstrations. However, traditional imitation learning methods require a large number of expert demonstrations in order to learn a complex behavior. Such a disadvantage has limited the potential of imitation learning in complex tasks where the expert demonstrations are not sufficient. In order to address the problem, we propose a Generative Adversarial Network-based model which is designed to learn optimal policies using only a single demonstration. The proposed model is evaluated on two simulated tasks in comparison with other methods. The results show that our proposed model is capable of completing co

Sufficient high-quality data are unavailable to describe the management approach and guideline of COVID-19 disease in pediatric and adolescent population which may be due to mild presentation in most of cases and less severe complications than older ages.

World Health Organization was concerned with the establishment of an approved guideline to manage the increasing number of COVID-19 patients worldwide aiming to prevent or lessen COVID-19 global burden.

The clinical features have a wide spectrum starting from uncomplicated mild illness, mild-moderate pneumonia, severe pneumonia, acute respiratory distress syndrome, sepsis, septic shock, and multisystem inflammatory syndrome in children.

Many important definitions

The communication networks (mobile phone networks, social media platforms) produce digital traces from their usages. This type of information help to understand and analyze the human mobility in very accurate way. By these analyzes over cities, it can give powerful data on daily citizen activities, urban planners have in that way, relevant indications for decision making on design and development. As well as, the Call detail Records (CDRs) provides valuable spatiotemporal data at the level of citywide or even nationwide. The CDRs could be analyzed to extract the life patterns and individuals mobility in an observed urban area and during ephemeral events. Whereas, their analysis gives conceptual views about human density and mobility pattern

In recent years, the performance of Spatial Data Infrastructures for governments and companies is a task that has gained ample attention. Different categories of geospatial data such as digital maps, coordinates, web maps, aerial and satellite images, etc., are required to realize the geospatial data components of Spatial Data Infrastructures. In general, there are two distinct types of geospatial data sources exist over the Internet: formal and informal data sources. Despite the growth of informal geospatial data sources, the integration between different free sources is not being achieved effectively. The adoption of this task can be considered the main advantage of this research. This article addresses the research question of how the