Federated And Transfer Learning

This book provides a collection of recent research works on learning from decentralized data, transferring information from one domain to another, and addressing theoretical issues on improving the privacy and incentive factors of federated learning as well as its connection with transfer learning and reinforcement learning. Over the last few years, the machine learning community has become fascinated by federated and transfer learning. Transfer and federated learning have achieved great success and popularity in many different fields of application. The intended audience of this book is students and academics aiming to apply federated and transfer learning to solve different kinds of real-world problems, as well as scientists, researchers, and practitioners in AI industries, autonomous vehicles, and cyber-physical systems who wish to pursue new scientific innovations and update their knowledge on federated and transfer learning and their applications.

The classic example of machine learning is based on isolated learning-a single model for each task using a single dataset. Most deep learning methods require a significant amount of labeled data, preventing their applicability in many areas where there is a shortage. In these cases, the ability of models to leverage information from unlabeled data or data that are not publicly available (for privacy and security reasons) can offer a remarkable alternative. Transfer learning and federated learning are alternative approaches that have emerged in recent years. More precisely, transfer learning is defined as the set of methods that leverage data from additional fields or tasks to train a model with greater generalizability and usually use a smaller amount of labeled data (via fine-tuning) to make them more specific for dedicated tasks. Accordingly, federated learning is a learning model that seeks to address the problem of data management and privacy through joint training with these data without the need to transfer the data to a central entity. With this in mind, this Special Issue of Applied Sciences provides an overview of the latest developments in this field.

This book shows how federated machine learning allows multiple data owners to collaboratively train and use a shared prediction model while keeping all the local training data private. Traditional machine learning approaches need to combine all data at one location, typically a data center, which may very well violate the laws on user privacy and data confidentiality. Today, many parts of the world demand that technology companies treat user data carefully according to user-privacy laws. The European Union's General Data Protection Regulation (GDPR) is a prime example. In this book, we describe how federated machine learning addresses this problem with novel solutions combining distributed machine learning, cryptography and security, and incentive mechanism design based on economic principles and game theory. We explain different types of privacy-preserving machine learning solutions and their technological backgrounds, and highlight some representative practical use cases. We show how federated learning can become the foundation of next-generation machine learning that caters to technological and societal needs for responsible AI development and application.

This in-depth tutorial for students, researchers, and developers covers foundations, plus applications ranging from search to multimedia.

This book covers the research area from multiple viewpoints including bibliometric analysis, reviews, empirical analysis, platforms, and future applications. The centralized training of deep learning and machine learning models not only incurs a high communication cost of data transfer into the cloud systems but also raises the privacy protection concerns of data providers. This book aims at targeting researchers and practitioners to delve deep into core issues in federated learning research to transform next-generation artificial intelligence applications. Federated learning enables the distribution of the learning models across the devices and systems which perform initial training and report the updated model attributes to the centralized cloud servers for secure and privacy-preserving attribute aggregation and global model development. Federated learning benefits in terms of privacy, communication efficiency, data security, and contributors’ control of their critical data.

Transfer learning is one of the most important technologies in the era of artificial intelligence and deep learning. It seeks to leverage existing knowledge by transferring it to another, new domain. Over the years, a number of relevant topics have attracted the interest of the research and application community: transfer learning, pre-training and fine-tuning, domain adaptation, domain generalization, and meta-learning. This book offers a comprehensive tutorial on an overview of transfer learning, introducing new researchers in this area to both classic and more recent algorithms. Most importantly, it takes a “student’s” perspective to introduce all the concepts, theories, algorithms, and applications, allowing readers to quickly and easily enter this area. Accompanying the book, detailed code implementations are provided to better illustrate the core ideas of several important algorithms, presenting good examples for practice.

This new book provides an in-depth understanding of federated learning, a new and increasingly popular learning paradigm that decouples data collection and model training via multi-party computation and model aggregation. The volume explores how federated learning integrates AI technologies, such as blockchain, machine learning, IoT, edge computing, and fog computing systems, allowing multiple collaborators to build a robust machine-learning model using a large dataset. It highlights the capabilities and benefits of federated learning, addressing critical issues such as data privacy, data security, data access rights, and access to heterogeneous data. The volume first introduces the general concepts of machine learning and then summarizes the federated learning system setup and its associated terminologies. It also presents a basic classification of FL, the application of FL for various distributed computing scenarios, an integrated view of applications of software-defined networks, etc. The book also explores the role of federated learning in the Internet of Medical Things systems as well. The book provides a pragmatic analysis of strategies for developing a communication-efficient federated learning system. It also details the applicability of blockchain with federated learning on IoT-based systems. It provides an in-depth study of FL-based intrusion detection systems, discussing their taxonomy and functioning and showcasing their superiority over existing systems. The book is unique in that it evaluates the privacy and security aspects in federated learning. The volume presents a comprehensive analysis of some of the common challenges, proven threats, and attack strategies affecting FL systems. Special coverage on protected shot-based federated learning for facial expression recognition is also included. This comprehensive book, Federated Learning: Principles, Paradigms, and Applications, will enable research scholars, information technology professionals, and distributed computing engineers to understand various aspects of federated learning concepts and computational techniques for real-life implementation.

Traditional books on machine learning can be divided into two groups- those aimed at advanced undergraduates or early postgraduates with reasonable mathematical knowledge and those that are primers on how to code algorithms. The field is ready for a text that not only demonstrates how to use the algorithms that make up machine learning methods, but