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.
Background: Impression materials, impression trays, and poured stone cast have been said to be the main source of cross infection between patients and dentists. However, it was observed that disinfection of the impression is not performed systematically in routine dental practice. Disinfection of alginates either by immersion or spray technique was found to cause dimensional inaccuracies, although with proper disinfection of alginates there were small dimensional changes. A variety of fluoride releasing products designed for topical use is currently available. Following their use, varied amount of fluoride is systemically absorbed depending on the fluoride concentration and the manner of its use. The objective of this study was to evaluate
... Show MoreDensity functional theory calculations are employed to investigate the impact of edifenphos molecule on the reactivity and electronic sensitivity of pure calcium oxide (CaO) nanocluster. The strong adsorption of edifenphos molecule on CaO nanocluster occurs by the sulfur head of the adsorbate, and the amount of the energy of this adsorption is around − 84.40 kcal/mol. The adsorption of edifenphos molecules results in a decrease in the values of Eg of CaO from 4.67 to 3.56 eV, as well as an increase in the electrical conductance. Moreover, the work function of CaO nanocluster is significantly affected, which changes the current of the field emission electron. Eventually, the recovery time is calculated around 99 ms at ambient temperature f
... Show MoreBackground and objectives: This study aimed at testing the effect of plastic sleeve or barrier, used to cover the guide of the light cure unit to prevent cross-infection, on the shear bond strength and site of bond failure of stainless steel and ceramic orthodontic brackets. Materials and methods: Forty orthodontic brackets; twenty stainless steel and twenty ceramic brackets bonded to forty extracted human maxillary first premolars using light cure adhesive cured with and without the use of a protective plastic barrier on the guide. Comparing the effect of this barrier on the shear bond strength and adhesive remnant index was performed using an independent t-test and Chi-square test. Results: The protective barrier had decreased the shear b
... Show MoreThis study investigates the influence of silver oxide (Ag2O) concentration on the optical characteristics of phosphate bioactive glasses (PBGs). PBGs have emerged as promising alternatives to conventional silicate glasses in the medical field due to their excellent bioactivity and chemical resistance. Samples with varying Ag2O concentrations (0, 0.25, 0.5, and 0.75g) were sintered at 780°C for 2 hrs in an electric furnace. The samples were subjected to Fourier transfer infrared spectroscopy (FTIR), ultraviolet-visible (UV-Vis) spectroscopy, and photoluminescence (PL) tests to assess their functional groups and optical properties. By analyzing the FTIR spectrum of phosphate bioactive glass containing different amounts of Ag2O, it is
... Show MoreIn this research, the electrical characteristics of glow discharge plasma were studied. Glow discharge plasma generated in a home-made DC magnetron sputtering system, and a DC-power supply of high voltage as input to the discharge electrodes were both utilized. The distance between two electrodes is 4cm. The gas used to produce plasma is argon gas which flows inside the chamber at a rate of 40 sccm. The influence of work function for different target materials (gold, copper, and silver), - 5cm in diameter and around 1mm thickness - different working pressures, and different applied voltages on electrical characteristics (discharge current, discharge potential, and Paschen’s curve) were studied. The results showed that the discharge cur
... Show MoreThe effect of molecules intersystem crossing (Kisc) on characteristics
(energy and duration) of a Passive Q- switched Laser Pulse has been
studied by mathematical description (rate equations model) for
temporal performance of which was used as a saturable absorber
material (passive switch) with laser. The study shows that the energy
and duration pulse are decreasing while the molecules intersystem
crossing into saturable absorber energy levels is increasing.
Kurdistan power system is expanded along years ago. The electrical power is transmitted through long transmission lines. The main problem of transmission lines is active and reactive power losses. It is important to solve this issue, unless, the most of electrical energy will lost over transmission system. In this study, High Voltage Direct Current links/bipolar connection were connected in a power system to reduce the power losses. The 132kV, 50 Hz, 36 buses Kurdistan power system is used as a study case. The load flow analysis was implemented by using ETAP.16 program in which Newton-Raphson method for three cases. The results show that the losses are reduced after inserted HVDC links.
User confidentiality protection is concerning a topic in control and monitoring spaces. In image, user's faces security in concerning with compound information, abused situations, participation on global transmission media and real-world experiences are extremely significant. For minifying the counting needs for vast size of image info and for minifying the size of time needful for the image to be address computationally. consequently, partial encryption user-face is picked. This study focuses on a large technique that is designed to encrypt the user's face slightly. Primarily, dlib is utilizing for user-face detection. Susan is one of the top edge detectors with valuable localization characteristics marked edges, is used to extract
... Show More