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.
يعد هذا النص أحد النصوص المسمارية المصادرة التي بحوزة المتحف العراقي، ويحمل الرقم المتحفي (235869)، قياساته )12،7x 6x 2،5سم). يتضمن مدخولات كميات من الشعير،أرخ النص الى عصر أور الثالثة (2012-2004 ق.م) و يعود الى السنة الثالثة من حكم الملك أبي-سين (2028-2004 ق.م)،أن الشخصية الرئيسة في هذا النص هو)با-اَ-كا مسمن الماشية( من مدينة أري-ساكرك، ومقارنته مع النصوص المسمارية المنشورة التي تعود الى أرشيفه يبلغ عددها (196) نصاً تضمنت نشاطاته م
... Show MoreAbstract
The research aims to build a training program to develop some executive functions for kindergarten children. To achieve this goal, the two researchers built the program according to the following steps:
1. Determining the general objective of the program.
2. Determining the behavioral objectives of the program.
3. Determining the included content in the program.
4. Implementing the content of the activities of the program.
5. Evaluating the Program.
The program included (12) training activities, the training activities included several items: the title of the activity, the time of implementation of the activity, the general objective of the activity, the procedural behavioral objective, the means and tools u
Praise be to Allah , the Lord of Heavens , Who revealed His Scripture in plain Arabic, and prayers and peace be upon the Master of all creatures and the chosen one , Muhammad (PBUH).Pushkin was known as a poet in Russia, the sun of its poetry that set, the father of the Russian literature and the founder of its literary language. When he published his poems, trying to express his inner poetic feelings and creative visions, he presented them with creativity and special touch. The reason behind the selection of Pushkin as the subject of this study is that he was affected by Islam and Arabs. For the purpose of this study, the dissertation has been divided into introduction and three chapters; each chapter includes two topics. In the introdu
... Show MorePDBN Rashid, International Journal of Development in Social Sciences and Humanities, 2023
In this paper, a modified derivation has been introduced to analyze the construction of C-space. The profit from using C-space is to make the process of path planning more safety and easer. After getting the C-space construction and map for two-link planar robot arm, which include all the possible situations of collision between robot parts and obstacle(s), the A* algorithm, which is usually used to find a heuristic path on Cartesian W-space, has been used to find a heuristic path on C-space map. Several modifications are needed to apply the methodology for a manipulator with degrees of freedom more than two. The results of C-space map, which are derived by the modified analysis, prove the accuracy of the overall C-space mapping and cons
... Show MoreSelf-repairing technology based on micro-capsules is an efficient solution for repairing cracked cementitious composites. Self-repairing based on microcapsules begins with the occurrence of cracks and develops by releasing self-repairing factors in the cracks located in concrete. Based on previous comprehensive studies, this paper provides an overview of various repairing factors and investigative methodologies. There has recently been a lack of consensus on the most efficient criteria for assessing self-repairing based on microcapsules and the smart solutions for improving capsule survival ratios during mixing. The most commonly utilized self-repairing efficiency assessment indicators are mechanical resistance and durab
... Show MorePorous materials play an important role in creating a sustainable environment by improving wastewater treatment's efficacy. Porous materials, including adsorbents or ion exchangers, catalysts, metal–organic frameworks, composites, carbon materials, and membranes, have widespread applications in treating wastewater and air pollution. This review examines recent developments in porous materials, focusing on their effectiveness for different wastewater pollutants. Specifically, they can treat a wide range of water contaminants, and many remove over 95% of targeted contaminants. Recent advancements include a wider range of adsorption options, heterogeneous catalysis, a new UV/H2O