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.
The parliamentary election is one of the features of democratic systems that give individuals the right to participate in government and political election-making. Typically, the process of parliamentary elections received wide attention from media, as well as attention from large segments of the public because they understand the vast importance to assume political positions and associated fates of people and their destinies. Its importance comes from the fact that it allows citizens the right to participate in managing the public affairs by granting their confidence and voices to the elected president or his representatives in the parliamentary.
Media task is to emerge democratic societies, in particular, in the mission of urging p
The increased applications of technology in the field of architecture, especially digital technology and aspects of automation, have made a major impact on various aspects of local architecture, especially the traditional ones. As these technologies have succeeded in integrating many technological applications in many traditional and heritage buildings and taking them to more complex uses. And included in it characteristics that were not contained, therefore the research problem was concentrated in the absence of a holistic view of the role of the aspects of automation as a technological and design effect and its mutual effects on traditional buildings (especially the traditional Bagh
Stereolithography (SLA) has become an essential photocuring 3D printing process for producing parts of complex shapes from photosensitive resin exposed to UV light. The selection of the best printing parameters for good accuracy and surface quality can be further complicated by the geometric complexity of the models. This work introduces multiobjective optimization of SLA printing of 3D dental bridges based on simple CAD objects. The effect of the best combination of a low-cost resin 3D printer’s machine parameter settings, namely normal exposure time, bottom exposure time and bottom layers for less dimensional deviation and surface roughness, was studied. A multiobjective optimization method was utilized, combining the Taguchi me
... Show MorePolish Academy of Sciences
Some coordination complexes of Co(ІІ), Ni(ІІ), Cu(ІІ), Cd(ІІ) and Hg(ІІ) are reacted in ethanol with Schiff base ligand derived from of 2,4,6- trihydroxybenzophenone and 3-aminophenol using microwave irradiation and then reacted with metal salts in ethanol as a solvent in 1:2 ratio (metal: ligand). The ligand [H4L] is characterized by FTIR, UV-Vis, C.H.N, 1H-NMR,13C-NMR, and mass spectra. The metal complexes are characterized by atomic absorption, infrared spectra, electronic spectra, molar conductance, (C.H.N for Ni(ІІ) complex) and magnetic moment measurements. These measurements indicate that the ligand coordinates with metal (ІІ) ion in a tridentate manner through the nitrogen and oxygen atoms of the ligand, octahed
... Show MoreSome coordination complexes of Co(??), Ni(??), Cu(??), Cd(??) and Hg(??) are reacted in ethanol with Schiff base ligand derived from of 2,4,6- trihydroxybenzophenone and 3-aminophenol using microwave irradiation and then reacted with metal salts in ethanol as a solvent in 1:2 ratio (metal: ligand). The ligand [H4L] is characterized by FTIR, UV-Vis, C.H.N, 1H-NMR,13C-NMR, and mass spectra. The metal complexes are characterized by atomic absorption, infrared spectra, electronic spectra, molar conductance, (C.H.N for Ni(??) complex) and magnetic moment measurements. These measurements indicate that the ligand coordinates with metal (??) ion in a tridentate manner through the nitrogen and oxygen atoms of the ligand, octahedral structures
... Show More