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
A surface fitting model is developed based on calorimeter data for two famous brands of household compressors. Correlation equations of ten coefficient polynomials were found as a function of refrigerant saturating and evaporating temperatures in range of (-35℃ to -10℃) using Matlab software for cooling capacity, power consumption, and refrigerant mass flow rate.
Additional correlations equations for these variables as a quick choice selection for a proper compressor use at ASHRAE standard that cover a range of swept volume range (2.24-11.15) cm3.
The result indicated that these surface fitting models are accurate with in ± 15% for 72 compressors model of cooling cap
... Show MoreThis paper presents an experimental and numerical study which was carried out to examine the influence of the size and the layout of the web openings on the load carrying capacity and the serviceability of reinforced concrete deep beams. Five full-scale simply supported reinforced concrete deep beams with two large web openings created in shear regions were tested up to failure. The shear span to overall depth ratio was (1.1). Square openings were located symmetrically relative to the midspan section either at the midpoint or at the interior boundaries of the shear span. Two different side dimensions for the square openings were considered, mainly, (200) mm and (230) mm. The strength results proved that the shear capacity of the dee
... Show MoreThis paper presents an experimental and numerical study which was carried out to examine the influence of the size and the layout of the web openings on the load carrying capacity and the serviceability of reinforced concrete deep beams. Five full-scale simply supported reinforced concrete deep beams with two large web openings created in shear regions were tested up to failure. The shear span to overall depth ratio was (1.1). Square openings were located symmetrically relative to the midspan section either at the midpoint or at the interior boundaries of the shear span. Two different side dimensions for the square openings were considered, mainly, (200) mm and (230) mm. The strength results proved that the shear capacity of the dee
... Show MoreModeling data acquisition systems (DASs) can support the vehicle industry in the development and design of sophisticated driver assistance systems. Modeling DASs on the basis of multiple criteria is considered as a multicriteria decision-making (MCDM) problem. Although literature reviews have provided models for DASs, the issue of imprecise, unclear, and ambiguous information remains unresolved. Compared with existing MCDM methods, the robustness of the fuzzy decision by opinion score method II (FDOSM II) and fuzzy weighted with zero inconsistency II (FWZIC II) is demonstrated for modeling the DASs. However, these methods are implemented in an intuitionistic fuzzy set environment that restricts the ability of experts to provide mem
... Show MorePolyacrylamide Solutions of different concentrations (0.2, 0.4, 0.6, 0.8, 1.0 %) of Ag nanoparticles and ZnO nanoparticles were prepared, the viscosities and surface tension were measured for all solutions, where measurements indicated an increase in these properties with increased concentration, where the relative viscosity of polyacrylamide/zinc nanoparticles increased from 1.275 to 2.243, and the relative viscosity of polyacrylamide/silver nanoparticles increased from 1.178 to 1.934. Viscosity is significant parameters during electrospinning process. While the surface tension of the polyacrylamide/zinc nanoparticles has changed from 0.0343 Nm-1 to .0.0.0 Nm-1 and changed from .0.000Nm-1 to.0.0.0 Nm-1. Also the constants KH and KK were
... Show MoreThis study aims to know the role of strategic leadership to achieving competitiveness in industrial establishments by identifying the respondents’ perceptions about the level of availability of dimensions of leadership strategies (creativity and innovation, risk tolerance, available opportunities) in Bashir Al-Siksek & Partners Company for the manufacture of sanitary and plastic ware in Gaza strip
To achieve this, a questionnaire was developed and distributed to a sample of managers, auditors, accountants, and administrative employees in the study sample company. The questionnaire tool was distributed to 60 employees and employees, of which (52) were retrieved, or 86.6%, and (8) were excluded for la
... Show MoreThe port of Gwadar is a major turning point in the Pakistani economy, as it is characterized by a very important strategic location and has characteristics that enable it to play an active role in the region, being a land bridge linking Central Asia, especially the countries of Central and South Asia, by the sea, and also linking the Chinese economy (west of China to the port). And due to its distinguished strategic location, which made it a center of attraction for large commercial ships, despite all the positive advantages, there are obstacles that stand in the way of completing it in the required manner represented by challenges (internal, regional and international) that fear the completion of the port, which will generate profit
... Show MoreIn this paper, image compression technique is presented based on the Zonal transform method. The DCT, Walsh, and Hadamard transform techniques are also implements. These different transforms are applied on SAR images using Different block size. The effects of implementing these different transforms are investigated. The main shortcoming associated with this radar imagery system is the presence of the speckle noise, which affected the compression results.