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.
Abstract
The Classical Normal Linear Regression Model Based on Several hypotheses, one of them is Heteroscedasticity as it is known that the wing of least squares method (OLS), under the existence of these two problems make the estimators, lose their desirable properties, in addition the statistical inference becomes unaccepted table. According that we put tow alternative, the first one is (Generalized Least Square) Which is denoted by (GLS), and the second alternative is to (Robust covariance matrix estimation) the estimated parameters method(OLS), and that the way (GLS) method neat and certified, if the capabilities (Efficient) and the statistical inference Thread on the basis of an acceptable
... Show MoreThe selection of proper field survey parameters of electrical resistivity can significantly provide efficient results within a reasonable time and cost. Four electrode arrays of 2D Electric Resistivity Imaging (ERI) surveys were applied to characterize and detect subsurface archaeological bodies and to determine the appropriate array type that should be applied in the field survey. This research is to identify the subsurface features of the Borsippa archaeological site, Babylon Governorate, Middle Iraq. Synthetic modeling studies were conducted to determine the proper array and parameters for imaging the shallow subsurface features or targets. The efficiency of many array types has been tested for the detection the buried archaeolog
... Show MoreMost of World nations are striving to provide the necessary needs to protect their economic properties assets against natural or abnormal disasters that may be inflicted on such property and the means that used by such countries to reduce the damages is insurance, whereas insurance as a system that collects and distributes different risks into the group thus to achieve a social symbiosis between individuals. The system works to transfer the risks from the individual to the group and then distributes the losses to all members of the group.
According to the importance of the insurance sector and the need to develop it as well as working on improving its performance, this search aims to identify the ac
... Show MoreThe aim of this study to identify the effect of using two strategies for active learning ( Jigsaw Strategy & Problems Solving) in learning some balanced beam's skills in artistic gymnastics for women , as well as to identify the best of the three methods (jigsaw strategy , problems solving and the traditional method) in learning some skills balance beam , the research has used the experimental methodology, and the subject included the students of the college of Physical Education and Sports Sciences / University of Baghdad / third grade and by the lot was selected (10) students for each group of groups Search three and The statistical package for social sciences (SPSS) was used means, the standard deviation and the (T.test), the one way a n
... Show MoreNonmissile penetrating spine injury (NMPSI) represents a small percent of spinal cord injuries (SCIs), estimated at 0.8% in Western countries. Regarding the causes, an NMPSI injury caused by a screwdriver is rare. This study reports a case of a retained double-headed screwdriver in a 37-year-old man who sustained a stab injury to the back of the neck, leaving the patient with a C4 Brown-Sequard syndrome (BSS). We discuss the intricacies of the surgical management of such cases with a literature review.
PubMed database was searched by the following combined formula of medical subjects headings,
This research deals with the design and simulation of a solar power system consisting of a KC200GT solar panel, a closed loop boost converter and a three phase inverter by using Matlab / Simulink. The mathematical equations of the solar panel design are presented. The electrical characteristics of the panel are tested at the values of 1000 for light radiation and 25 °C for temperature environment. The Proportional Integral (PI) controller is connected as feedback with the Boost converter to obtain a stable output voltage by reducing the oscillations in the voltage to charge a battery connected to the output of the converter. Two methods (Particle Swarm Optimization (PSO) and Zeigler- Nichols) are used for tuning
... Show Moren this study, data or X-ray images Fixable Image Transport System (FITS) of objects were analyzed, where energy was collected from the body by several sensors; each sensor receives energy within a specific range, and when energy was collected from all sensors, the image was formed carrying information about that body. The images can be transferred and stored easily. The images were analyzed using the DS9 program to obtain a spectrum for each object,an energy corresponding to the photons collected per second. This study analyzed images for two types of objects (globular and open clusters). The results showed that the five open star clusters contain roughly t
... Show MoreWith the revolutionized expansion of the Internet, worldwide information increases the application of communication technology, and the rapid growth of significant data volume boosts the requirement to accomplish secure, robust, and confident techniques using various effective algorithms. Lots of algorithms and techniques are available for data security. This paper presents a cryptosystem that combines several Substitution Cipher Algorithms along with the Circular queue data structure. The two different substitution techniques are; Homophonic Substitution Cipher and Polyalphabetic Substitution Cipher in which they merged in a single circular queue with four different keys for each of them, which produces eight different outputs for
... Show More