Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.
This research aims to examine the relationship between learning organization and behavior of work teams. The variable of the learning organization took four dimensions depending on the study (sudhartna & Li, 2004): Common cultural values , communication, knowledge transfer and the characteristics of workers. The behavior of teams was identified on the basis of realizing of the respondents of their organization to work as a team where the research relied concepts applied in the study (Hakim , 2005) , and chose to research the case of a service organization for the study and relied on four dimensions of coordination , cooperation , sharing of information , the performance of the team, and was a curriculum approach and des
... Show MoreThe adopted method in the teaching of history is conservation and indoctrination in all grades, and this will lead to a lack of students interact with teachers in the course of the lesson, and poor use of teachers to questions that raise students' thinking during the lesson, which leads to a lack of interest in the topic of the lesson and wasting opportunities contribution making it the teacher at the center of the educational process, and to provide arrogating the researcher to contribute to teaching style with the belief that the use of this method of teaching could lead to overcome the difficulties and problems faced by the teaching material.
And there are educational complexes integrated approac
... Show MoreThe current paper aims to identify potential factors associated with employees’ intentions to leave information and communication technology companies in Iraq. There is evident variability in the literature regarding these factors; hence, a factor analysis approach was employed to identify these factors within the surveyed environment. Due to the difficulty in precisely delineating the size of the research population, a purposive sampling method was employed to reach an appropriate number of respondents within the aforementioned companies. A total of 288 employees responded to the survey conducted via Google Forms. The test results revealed the presence of five primary factors associated with employees’ intentions to leave, name
... Show MoreIn this paper, chip and powder copper are used as reinforcing phase in polyester matrix to form composites. Mechanical properties such as flexural strength and impact test of polymer reinforcement copper (powder and chip) were done, the maximum flexural strength for the polymer reinforcement with copper (powder and chip) are (85.13 Mpa) and (50.08 Mpa) respectively was obtained, while the maximum observation energy of the impact test for the polymer reinforcement with copper (powder and chip) are (0.85 J) and (0.4 J) respectively
In this paper the experimentally obtained conditions for the fusion splicing with photonic crystal fibers (PCF) having large mode areas were reported. The physical mechanism of the splice loss and the microhole collapse property of photonic crystal fiber (PCF) were studied. By controlling the arc-power and the arc-time of a conventional electric arc fusion splicer (FSM-60S), the minimum loss of splicing for fusion two conventional single mode fibers (SMF-28) was (0.00dB), which has similar mode field diameter. For splicing PCF (LMA-10) with a conventional single mode fiber (SMF-28), the loss was increased due to the mode field mismatch.
Double-layer micro-perforated panels (MPPs) have been studied extensively as sound absorption systems to increase the absorption performance of single-layer MPPs. However, existing proposed models indicate that there is still room for improvement regarding the frequency bands of absorption for the double-layer MPP. This study presents a double-layer MPP formed with two single MPPs with inhomogeneous perforation backed by multiple cavities of varying depths. The theoretical formulation is developed using the electrical equivalent circuit method to calculate the absorption coefficient under a normal incident sound. The simulation results show that the proposed model can produce absorption coefficient with wider absorption bandwidth compared w
... Show More