Page 55
Page 55
Deepfake detection = اكتشاف التزييف العميق
In the rapidly evolving era of artificial intelligence, addressing the escalating threats of deepfake technology becomes a necessity because of the increasing sophistication of AI algorithms in generating deceptive content, and since it threatens the integrity of information across diverse data. The main objective is to build a sophisticated AI-driven system to detect different types of deepfake in text, audio, and images. In English text deepfake detection, multiple pre-trained tokenizers have been used, but XLNET and BERT stand out with identifying objects outside the dataset with an accuracy of 0.9809 and both have been generalized & trained using LSTM. In Arabic text deepfake detection, Arabert has been trained using LSTM which led with an accuracy of 99.53% by generalizing the model. Both English and Arabic datasets have been generated to enhance the accuracy and effectiveness of the models. Audio deepfake detection has been generalized too, using Random Forest with an accuracy of 98.259%.
Deep Statistical Comparison for Meta-heuristic Stochastic Optimization Algorithms
Presents a comprehensive comparison of the performance of stochastic optimization algorithms / Includes an introduction to benchmarking and statistical analysis / Provides a web-based tool for making statistical comparisons of optimization algorithms / Overviews of the current approaches used to analyze algorithm performance in a range of common scenarios, while also addressing issues that are often overlooked. In turn, it shows how these issues can be easily avoided by applying the principles that have produced Deep Statistical Comparison and its variants. The focus is on statistical analyses performed using single-objective and multi-objective optimization data. At the end of the book, examples from a recently developed web-service-based e-learning tool (DSCTool) are presented. The tool provides users with all the functionalities needed to make robust statistical comparison analyses in various statistical scenarios. The book is intended for newcomers to the field and experienced researchers alike. For newcomers, it covers the basics of optimization and statistical analysis, familiarizing them with the subject matter before introducing the Deep Statistical Comparison approach. Experienced researchers can quickly move on to the content on new statistical approaches.
Deep neural networks and data for automated driving : robustness, uncertainty quantification, and insights towards safety
Environment perception for highly automated driving heavily employs deep neural networks, facing many challenges. How much data do we need for training and testing? How to use synthetic data to save labeling costs for training? How do we increase robustness and decrease memory usage? For inevitably poor conditions: How do we know that the network is uncertain about its decisions? Can we understand a bit more about what actually happens inside neural networks? This leads to a very practical problem particularly for DNNs employed in automated driving: What are useful validation techniques and how about safety? This book unites the views from both academia and industry, where computer vision and machine learning meet environment perception for highly automated driving. Naturally, aspects of data, robustness, uncertainty quantification, and, last but not least, safety are at the core of it. This book is unique: In its first part, an extended survey of all the relevant aspects is provided. The second part contains the detailed technical elaboration of the various questions mentioned above.
Deep Learning, Machine Learning and IoT in Biomedical and Health Informatics : Techniques and Applications
Examines and demonstrates state-of-the-art approaches for IoT and Machine Learning based biomedical and health related applications. This book aims to provide computational methods for accumulating, updating and changing knowledge in intelligent systems and particularly learning mechanisms that help us to induce knowledge from the data. It is helpful in cases where direct algorithmic solutions are unavailable, there is lack of formal models, or the knowledge about the application domain is inadequately defined. In the future IoT has the impending capability to change the way we work and live. These computing methods also play a significant role in design and optimization in diverse engineering disciplines. With the influence and the development of the IoT concept, the need for AI (artificial intelligence) techniques has become more significant than ever.
Deep Learning with PyTorch Lightning : Build and train high-performance artificial intelligence and self-supervised models using Python
You’ll learn how to configure PyTorch Lightning on a cloud platform, understand the architectural components, and explore how they are configured to build various industry solutions. You’ll build a neural network architecture, deploy an application from scratch, and see how you can expand it based on your specific needs, beyond what the framework can provide. In the later chapters, you’ll also learn how to implement capabilities to build and train various models like Convolutional Neural Nets (CNN), Natural Language Processing (NLP), Time Series, Self-Supervised Learning, Semi-Supervised Learning, Generative Adversarial Network (GAN) using PyTorch Lightning.
Deep learning pipeline : Building a deep learning model with TensorFlow
Build your own pipeline based on modern TensorFlow approaches rather than outdated engineering concepts. This book shows you how to build a deep learning pipeline for real-life TensorFlow projects. You'll learn what a pipeline is and how it works so you can build a full application easily and rapidly. Then troubleshoot and overcome basic Tensorflow obstacles to easily create functional apps and deploy well-trained models. Step-by-step and example-oriented instructions help you understand each step of the deep learning pipeline while you apply the most straightforward and effective tools to demonstrative problems and datasets.
Deep learning architecture and application
As one of the fastest-growing topics in machine learning, deep learning algorithms have achieved unprecedented success in recent years. Novel paradigms (such as contrastive learning and few-shot learning) in deep learning and rising neural network architectures (e.g., transformer and masked autoencoder) are dramatically changing the field of data-driven algorithms. More importantly, deep learning models are redefining the next generation of industrial applications spanning image recognition, speech processing, language translation, healthcare, and other sciences. For example, recent advances in deep representation learning are allowing us to learn about protein 3D structures, which sheds new light on fundamental medicine and biology along with potentially bringing in billions of dollars (e.g., in the pharmaceutical market).
Deep learning and computer vision in remote sensing-II
Computer vision (CV) have seen a massive rise in popularity in the remote sensing field over the last few years. This success is mostly due to the effectiveness of deep learning (DL) algorithms. However, remote sensing data acquisition and annotation, as well as information extraction from massive remote sensing data, are still challenging. This reprint collected novel developments in the field of deep learning and computer vision methods for remote sensing. Papers dealing with fundamental theoretical analyses, as well as those demonstrating their application to real-world problems, have been published. With practical examples and real-world case studies, this reprint provides a valuable resource for researchers, professionals, and students seeking to harness the power of deep learning in the field of remote sensing.
Deep learning and computer vision in remote sensing-I
In the last few years, huge amounts of progress have been made regarding remote sensing in the field of computer vision. This success and progress is mostly due to the effectiveness of deep learning (DL) algorithms. In addition, the remote sensing community has shifted its attention to DL, and DL algorithms have been used to achieve significant success in many image analysis tasks. However, with regard to remote sensing, a number of challenges caused by difficulties in data acquisition and annotation have not been fully solved yet. This reprint is a collection of novel developments in the field of remote sensing using computer vision, deep learning, and artificial intelligence. The articles published involve fundamental theoretical analyses as well as those demonstrating their application to real-world problems.
Deep Integration, Global Firms, and Technology Spillovers
This book explores the impact of deep regional economic integration on spillovers of knowledge and technology across countries. Deep integration through signing deep regional trade agreements (DRTAs), which cover various policy areas in addition to tariff reductions, may or may not facilitate technology spillovers among their signatories. To understand the mechanism of the impact of deep integration on technology spillovers, this book starts by analyzing the behavior of global firms. Factors that affect global firms’ activities, such as export, foreign direct investment (FDI), offshore outsourcing, are examined. Micro data on Japanese firms are employed for the analysis. Then, the relationships between bilateral trade patterns and technology spillovers and between types of FDI and technology spillovers are investigated in detail.
Deep impact mission : Looking beneath the surface of a cometary nucleus
Deep Impact, or at least part of the flight system, is designed to crash into comet 9P/Tempel 1. This bold mission design enables cometary researchers to peer into the cometary nucleus, analyzing the material excavated with its imagers and spectrometers. The book describes the mission, its objectives, expected results, payload, and data products in articles written by those most closely involved. This mission has the potential of revolutionizing our understanding of the cometary nucleus.
Deep fake detection
Deep learning has been successfully applied to solve various complex problems ranging from big data analytics to computer vision and human-level control. Deep learning advances however have also been employed to create software that can cause threats to privacy, democracy and national security. One of those deep learning-powered applications recently emerged is “deepfake”. Deepfake algorithms can create fake images and videos that humans cannot distinguish them from authentic ones. The proposal of technologies that can automatically detect and assess the integrity of digital visual media is therefore indispensable.
Deep Data Analytics for New Product Development
The benefits of reading this book are twofold. The first is an understanding of the stages of a new product development process from ideation through launching and tracking, each supported by information about customers. The second benefit is an understanding of the deep data analytics for extracting that information from data. These analytics, drawn from the statistics, econometrics, market research, and machine learning spaces, are developed in detail and illustrated at each stage of the process with simulated data. The stages of new product development and the supporting deep data analytics at each stage are not presented in isolation of each other, but are presented as a synergistic whole.
Decrypted Secrets : Methods and Maxims of Cryptology
Cryptology, for millennia a "secret science", is rapidly gaining in practical importance for the protection of communication channels, databases, and software. Beside its role in computerized information systems (public key systems), more and more applications within computer systems and networks are appearing, which also extend to access rights and source file protection. The first part of this book treats secret codes and their uses - cryptography. The second part deals with the process of covertly decrypting a secret code - cryptanaly-sis - where in particular advice on assessing methods is given. The book presupposes only elementary mathematical knowledge.
Decoding the city urbanism in the Age of Big Data
Shows how Big Data change reality and, hence, the way we deal with the city. They demonstrate how the Lab interprets digital data as material that can be used for the formulation of a different urban future. The publication also looks at the negative aspects of the city-related data acquisition and control.
Declarative programming for knowledge management ; 16th International conference on applications of declarative programming and knowledge management, INAP 2005, Fukuoka, Japan, October 22-24, 2005. Revised Selected Papers
Presents a selection of papers presented at the 16th Inter- tional Conference on Applications of Declarative Programming and Knowledge Management, INAP 2005,held in October 2005 at Waseda University, Fukuoka, Japan. These papers re?ect a snapshot of ongoing research and current app- cations in knowledge management and declarative programming.
Datatype-Generic Programming ; International Spring School, SSDGP 2006, Nottingham, UK, April 24-27, 2006, Revised Lectures
A leitmotif in the evolution of programming paradigms has been the level and extent of parametrisation that is facilitated — the so-called genericity of the paradigm. The sorts of parameters that can be envisaged in a programming language range from simple values, like integers and fioating-point numbers, through structured values, types and classes, to kinds (the type of types and/or classes).Datatype-generic programming is about parametrising programsby the structure of the data that they manipulate. To appreciate the importance of data type genericity,one need look no further than the internet. The internet is a massive repository of structured data, but the structure is rarely exploited. For example, compression of data can be much more efiective if its structure is known, but most compression algorithms regard the input data as simply a string of bits, and take no account of its internal organisation. Datatype-generic programming is about exploiting the structure of data when it is relevant and ignoring it when it is not. Programming languages most c- monly used at the present time do not provide efiective mechanisms for do- menting and implementing datatype genericity.
Dataset Studio
Data is the new oil, which means that AI engineers can face difficulties in locating suitable datasets. Dataset Studio is a comprehensive platform designed to support AI engineers in the creation and optimization of datasets. This project offers a diverse range of services that encompass data collection, data augmentation, and data classification. As a result, this software empowers engineers by automatically generating structured data through the utilization of advanced tools and AI techniques. By automating the laborious tasks of manual data collection and extraction, Dataset Studio effectively streamlines the workflow for AI engineers, enabling them to save valuable time and focus on the more intricate aspects of dataset development and refinement.
Data-Driven Policy Impact Evaluation : How Access to Microdata is Transforming Policy Design
Provides statistical tools for evaluating the effects of public policies advocated by governments and public institutions. Experts from academia, national statistics offices and various research centers present modern econometric methods for an efficient data-driven policy evaluation and monitoring, assess the causal effects of policy measures and report on best practices of successful data management and usage. Topics include data confidentiality, data linkage, and national practices in policy areas such as public health, education and employment. It offers scholars as well as practitioners from public administrations, consultancy firms and nongovernmental organizations insights into counterfactual impact evaluation methods and the potential of data-based policy and program evaluation.
Data-Driven Fault Detection and Reasoning for Industrial Monitoring
Assesses the potential of data-driven methods in industrial process monitoring engineering. The process modeling, fault detection, classification, isolation, and reasoning are studied in detail. These methods can be used to improve the safety and reliability of industrial processes. Fault diagnosis, including fault detection and reasoning, has attracted engineers and scientists from various fields such as control, machinery, mathematics, and automation engineering. Combining the diagnosis algorithms and application cases, this book establishes a basic framework for this topic and implements various statistical analysis methods for process monitoring.
