Lifelong learning addresses situations in which a learner faces a series of different learning tasks providing the opportunity for synergy among them. Explanation-based neural network learning (EBNN) is a machine learning algorithm that transfers knowledge across multiple learning tasks. When faced with a new learning task, EBNN exploits domain knowledge accumulated in previous learning tasks to guide generalization in the new one. As a result, EBNN generalizes more accurately from less data than comparable methods. Explanation-Based Neural Network Learning: A Lifelong Learning Approach describes the basic EBNN paradigm and investigates it in the context of supervised learning, reinforcement learning, robotics, and chess. The paradigm of lifelong learning - using earlier learned knowledge to improve subsequent learning - is a promising direction for a new generation of machine learning algorithms. Given the need for more accurate learning methods, it is difficult to imagine a future for machine learning that does not include this paradigm.' From the Foreword by Tom M. Mitchell.

This book is intended to assist in the development of smart and efficient green energy solutions. It introduces energy systems, power generation, and power demands which able to minimise generation costs, power loss or environmental effects. It proposes cutting-edge solutions and approaches based on recent technologies such as intelligent renewable energy systems (wind and solar). These solutions, applied to different sectors, can provide a solid basis for meeting the needs of both developed and developing countries. The book provides a collection of contributions including new techniques, methods, algorithms, practical solutions and models based on applying artificial intelligence and the Internet of things into green energy management systems. It provides a comprehensive reference for researchers, scholars and industry in the field of green energy and computational intelligence.

This volume offers a wealth of interdisciplinary approaches to artificial intelligence, machine learning and optimization tools, which contribute to the optimization of urban features towards forming smart, sustainable, and livable future cities. Special features include: New research on the design of city elements and smart systems with respect to new technologies and scientific thinking Discussions on the theoretical background that lead to smart cities for the future New technologies and principles of research that can promote ideas of artificial intelligence and machine learning in optimized urban environments The book engages students and researchers in the subjects of artificial intelligence, machine learning, and optimization tools in smart sustainable cities as eminent international experts contribute their research results and thinking in its chapters. Overall, its audience can benefit from a variety of disciplines including, architecture, engineering, physics, mathematics, computer science, and related fields.

Computational Architectures Integrating Neural and Symbolic Processes: A Perspective on the State of the Art focuses on a currently emerging body of research. With the reemergence of neural networks in the 1980s with their emphasis on overcoming some of the limitations of symbolic AI, there is clearly a need to support some form of high-level symbolic processing in connectionist networks. As argued by many researchers, on both the symbolic AI and connectionist sides, many cognitive tasks, e.g. language understanding and common sense reasoning, seem to require high-level symbolic capabilities. How these capabilities are realized in connectionist networks is a difficult question and it constitutes the focus of this book. Computational Architectures Integrating Neural and Symbolic Processes addresses the underlying architectural aspects of the integration of neural and symbolic processes. In order to provide a basis for a deeper understanding of existing divergent approaches and provide insight for further developments in this field, this book presents: (1) an examination of specific architectures (grouped together according to their approaches), their strengths and weaknesses, why they work, and what they predict, and (2) a critique/comparison of these approaches. Computational Architectures Integrating Neural and Symbolic Processes is of interest to researchers, graduate students, and interested laymen, in areas such as cognitive science, artificial intelligence, computer science, cognitive psychology, and neurocomputing, in keeping up-to-date with the newest research trends. It is a comprehensive, in-depth introduction to this new emerging field.

With approximately 2500 problems, this book provides a collection of practical problems on the basic and advanced data structures, design, and analysis of algorithms. To make this book suitable for self-instruction, about one-third of the algorithms are supported by solutions, and some others are supported by hints and comments. This book is intended for students wishing to deepen their knowledge of algorithm design in an undergraduate or beginning graduate class on algorithms, for those teaching courses in this area, for use by practicing programmers who wish to hone and expand their skills, and as a self-study text for graduate students who are preparing for the qualifying examination on algorithms for a Ph.D. program in Computer Science or Computer Engineering. About all, it is a good source for exam problems for those who teach algorithms and data structure. The format of each chapter is just a little bit of instruction followed by lots of problems. This book is intended to augment the problem sets found in any standard algorithms textbook. This book * begins with four chapters on background material that most algorithms instructors would like their students to have mastered before setting foot in an algorithms class. The introductory chapters include mathematical induction, complexity notations, recurrence relations, and basic algorithm analysis methods. * provides many problems on basic and advanced data structures including basic data structures (arrays, stack, queue, and linked list), hash, tree, search, and sorting algorithms. * provides many problems on algorithm design techniques: divide and conquer, dynamic programming, greedy algorithms, graph algorithms, and backtracking algorithms. * is rounded out with a chapter on NP-completeness.

This book presents a variety of techniques designed to enhance and empower multi-disciplinary and multi-institutional machine learning research in healthcare informatics. It is intended to provide a unique compendium of current and emerging machine learning paradigms for healthcare informatics, reflecting the diversity, complexity, and depth and breadth of this multi-disciplinary area.

This book investigates in detail the deep learning (DL) techniques in electromagnetic (EM) near-field scattering problems, assessing its potential to replace traditional numerical solvers in real-time forecast scenarios. Studies on EM scattering problems have attracted researchers in various fields, such as antenna design, geophysical exploration and remote sensing. Pursuing a holistic perspective, the book introduces the whole workflow in utilizing the DL framework to solve the scattering problems. To achieve precise approximation, medium-scale data sets are sufficient in training the proposed model. As a result, the fully trained framework can realize three orders of magnitude faster than the conventional FDFD solver. It is worth noting that the 2D and 3D scatterers in the scheme can be either lossless medium or metal, allowing the model to be more applicable. This book is intended for graduate students who are interested in deep learning with computational electromagnetics, professional practitioners working on EM scattering, or other corresponding researchers.

With the field of computational statistics growing rapidly, there is a need for capturing the advances and assessing their impact. Advances in simulation and graphical analysis also add to the pace of the statistical analytics field. Computational statistics play a key role in financial applications, particularly risk management and derivative pricing, biological applications including bioinformatics and computational biology, and computer network security applications that touch the lives of people. With high impacting areas such as these, it becomes important to dig deeper into the subject and explore the key areas and their progress in the recent past. Methodologies and Applications of Computational Statistics for Machine Intelligence serves as a guide to the applications of new advances in computational statistics. This text holds an accumulation of the thoughts of multiple experts together, keeping the focus on core computational statistics that apply to all domains. Covering topics including artificial intelligence, deep learning, and trend analysis, this book is an ideal resource for statisticians, computer scientists, mathematicians, lecturers, tutors, researchers, academic and corporate libraries, practitioners, professionals, students, and academicians.

This book focuses on the fundamentals of deep learning along with reporting on the current state-of-art research on deep learning. In addition, it provides an insight of deep neural networks in action with illustrative coding examples. Deep learning is a new area of machine learning research which has been introduced with the objective of moving ML closer to one of its original goals, i.e. artificial intelligence. Deep learning was developed as an ML approach to deal with complex input-output mappings. While traditional methods successfully solve problems where final value is a simple function of input data, deep learning techniques are able to capture composite relations between non-immediately related fields, for example between air pressure recordings and English words, millions of pixels and textual description, brand-related news and future stock prices and almost all real world problems. Deep learning is a class of nature inspired machine learning algorithms that uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The learning may be supervised (e.g. classification) and/or unsupervised (e.g. pattern analysis) manners. These algorithms learn multiple levels of representations that correspond to different levels of abstraction by resorting to some form of gradient descent for training via backpropagation. Layers that have been used in deep learning include hidden layers of an artificial neural network and sets of propositional formulas. They may also include latent variables organized layer-wise in deep generative models such as the nodes in deep belief networks and deep boltzmann machines. Deep learning is part of state-of-the-art systems in various disciplines, particularly computer vision, automatic speech recognition (ASR) and human action recognition.

This comprehensive and timely book, New Age Analytics: Transforming the Internet through Machine Learning, IoT, and Trust Modeling, explores the importance of tools and techniques used in machine learning, big data mining, and more. The book explains how advancements in the world of the web have been achieved and how the experiences of users can be analyzed. It looks at data gathering by the various electronic means and explores techniques for analysis and management, how to manage voluminous data, user responses, and more. This volume provides an abundance of valuable information for professionals and researchers working in the field of business analytics, big data, social network data, computer science, analytical engineering, and forensic analysis. Moreover, the book provides insights and support from both practitioners and academia in order to highlight the most debated aspects in the field.

This book presents an exciting new synthesis of directed and undirected, discrete and continuous graphical models. Combining elements of Bayesian networks and Markov random fields, the newly introduced hybrid random fields are an interesting approach to get the best of both these worlds, with an added promise of modularity and scalability. The authors have written an enjoyable book---rigorous in the treatment of the mathematical background, but also enlivened by interesting and original historical and philosophical perspectives. -- Manfred Jaeger, Aalborg Universitet The book not only marks an effective direction of investigation with significant experimental advances, but it is also---and perhaps primarily---a guide for the reader through an original trip in the space of probabilistic modeling. While digesting the book, one is enriched with a very open view of the field, with full of stimulating connections. ...] Everyone specifically interested in Bayesian networks and Markov random fields should not miss it. -- Marco Gori, Universita degli Studi di Siena Graphical models are sometimes regarded---incorrectly---as an impractical approach to machine learning, assuming that they only work well for low-dimensional applications and discrete-valued domains. While guiding the reader through the major achievements of this research area in a technically detailed yet accessible way, the book is concerned with the presentation and thorough (mathematical and experimental) investigation of a novel paradigm for probabilistic graphical modeling, the hybrid random field. This model subsumes and extends both Bayesian networks and Markov random fields. Moreover, it comes with well-defined learning algorithms, both for discrete and continuous-valued domains, which fit the needs of real-world applications involving large-scale, high-dimensional data.

This book is written both for readers entering the field, and for practitioners with a background in AI and an interest in developing real-world applications. The book is a great resource for practitioners and researchers in both industry and academia, and the discussed case studies and associated material can serve as inspiration for a variety of projects and hands-on assignments in a classroom setting. I will certainly keep this book as a personal resource for the courses I teach, and strongly recommend it to my students. --Dr. Carlotta Domeniconi, Associate Professor, Computer Science Department, GMU This book offers a curriculum for introducing interpretability to machine learning at every stage. The authors provide compelling examples that a core teaching practice like leading interpretive discussions can be taught and learned by teachers and sustained effort. And what better way to strengthen the quality of AI and Machine learning outcomes. I hope that this book will become a primer for teachers, data Science educators, and ML developers, and together we practice the art of interpretive machine learning. --Anusha Dandapani, Chief Data and Analytics Officer, UNICC and Adjunct Faculty, NYU This is a wonderful book! I'm pleased that the next generation of scientists will finally be able to learn this important topic. This is the first book I've seen that has up-to-date and well-rounded coverage. Thank you to the authors! --Dr. Cynthia Rudin, Professor of Computer Science, Electrical and Computer Engineering, Statistical Science, and Biostatistics & Bioinformatics Literature on Explainable AI has up until now been relatively scarce and featured mainly mainstream algorithms like SHAP and LIME. This book has closed this gap by providing an extremely broad review of various algorithms proposed in the scientific circles over the previous 5-10 years. This book is a great guide to anyone who is new to the field of XAI or is already familiar with the field and is willing to expand their knowledge. A comprehensive review of the state-of-the-art Explainable AI methods starting from visualization, interpretable methods, local and global explanations, time series methods, and finishing with deep learning provides an unparalleled source of information currently unavailable anywhere else. Additionally, notebooks with vivid examples are a great supplement that makes the book even more attractive for practitioners of any level. Overall, the authors provide readers with an enormous breadth of coverage without losing sight of practical aspects, which makes this book truly unique and a great addition to the library of any data scientist. Dr. Andrey Sharapov, Product Data Scientist, Explainable AI Expert and Speaker, Founder of Explainable AI-XAI Group

Deep Learning models are at the core of artificial intelligence research today. It is well known that deep learning techniques are disruptive for Euclidean data, such as images or sequence data, and not immediately applicable to graph-structured data such as text. This gap has driven a wave of research for deep learning on graphs, including graph representation learning, graph generation, and graph classification. The new neural network architectures on graph-structured data (graph neural networks, GNNs in short) have performed remarkably on these tasks, demonstrated by applications in social networks, bioinformatics, and medical informatics. Despite these successes, GNNs still face many challenges ranging from the foundational methodologies to the theoretical understandings of the power of the graph representation learning. This book provides a comprehensive introduction of GNNs. It first discusses the goals of graph representation learning and then reviews the history, current developments, and future directions of GNNs. The second part presents and reviews fundamental methods and theories concerning GNNs while the third part describes various frontiers that are built on the GNNs. The book concludes with an overview of recent developments in a number of applications using GNNs. This book is suitable for a wide audience including undergraduate and graduate students, postdoctoral researchers, professors and lecturers, as well as industrial and government practitioners who are new to this area or who already have some basic background but want to learn more about advanced and promising techniques and applications.

This book takes the notions of adaptivity and learning from the realm of engineering into the realm of biology and natural processes. It introduces a Hebbian-LMS algorithm, an integration of unsupervised Hebbian learning and supervised LMS learning in neural networks, as a mathematical representation of a general theory for synaptic learning in the brain, and adaptation and functional control of homeostasis in living systems. Written in a language that is able to address students and scientists with different backgrounds, this book accompanies readers on a unique journey through various homeostatic processes in living organisms, such as body temperature control and synaptic plasticity, explaining how the Hebbian-LMS algorithm can help understand them, and suggesting some open questions for future research. It also analyses cell signalling pathways from an unusual perspective, where hormones and hormone receptors are shown to be regulated via the principles of the Hebbian-LMS algorithm. It further discusses addiction and pain, and various kinds of mood disorders alike, showing how they can be modelled with the Hebbian-LMS algorithm. For the first time, the Hebbian-LMS algorithm, which has been derived from a combination of Hebbian theory from the neuroscience field and the LMS algorithm from the engineering field of adaptive signal processing, becomes a potent model for understanding how biological regulation works. Thus, this book is breaking new ground in neuroscience by providing scientists with a general theory for how nature does control synaptic learning. It then goes beyond that, showing that the same principles apply to hormone-mediated regulation of physiological processes. In turn, the book tackles in more depth the concept of learning. It covers computer simulations and strategies for training neural networks with the Hebbian-LMS algorithm, demonstrating that the resulting algorithms are able to identify relationships between unknown input patterns. It shows how this can translate in useful ideas to understand human memory and design cognitive structures. All in all, this book offers an absolutely, unique, inspiring reading for biologists, physiologists, and engineers, paving the way for future studies on what we could call the nature's secret learning algorithm.

arise automatically as a result of the recursive structure of the task and the continuous nature of the SRN's state space. Elman also introduces a new graphical technique for study ing network behavior based on principal components analysis. He shows that sentences with multiple levels of embedding produce state space trajectories with an intriguing self similar structure. The development and shape of a recurrent network's state space is the subject of Pollack's paper, the most provocative in this collection. Pollack looks more closely at a connectionist network as a continuous dynamical system. He describes a new type of machine learning phenomenon: induction by phase transition. He then shows that under certain conditions, the state space created by these machines can have a fractal or chaotic structure, with a potentially infinite number of states. This is graphically illustrated using a higher-order recurrent network trained to recognize various regular languages over binary strings. Finally, Pollack suggests that it might be possible to exploit the fractal dynamics of these systems to achieve a generative capacity beyond that of finite-state machines."

The transition towards exascale computing has resulted in major transformations in computing paradigms. The need to analyze and respond to such large amounts of data sets has led to the adoption of machine learning (ML) and deep learning (DL) methods in a wide range of applications. One of the major challenges is the fetching of data from computing memory and writing it back without experiencing a memory-wall bottleneck. To address such concerns, in-memory computing (IMC) and supporting frameworks have been introduced. In-memory computing methods have ultra-low power and high-density embedded storage. Resistive Random-Access Memory (ReRAM) technology seems the most promising IMC solution due to its minimized leakage power, reduced power consumption and smaller hardware footprint, as well as its compatibility with CMOS technology, which is widely used in industry. In this book, the authors introduce ReRAM techniques for performing distributed computing using IMC accelerators, present ReRAM-based IMC architectures that can perform computations of ML and data-intensive applications, as well as strategies to map ML designs onto hardware accelerators. The book serves as a bridge between researchers in the computing domain (algorithm designers for ML and DL) and computing hardware designers.

With an evolutionary advancement of Machine Learning (ML) algorithms, a rapid increase of data volumes and a significant improvement of computation powers, machine learning becomes hot in different applications. However, because of the nature of "black-box" in ML methods, ML still needs to be interpreted to link human and machine learning for transparency and user acceptance of delivered solutions. This edited book addresses such links from the perspectives of visualisation, explanation, trustworthiness and transparency. The book establishes the link between human and machine learning by exploring transparency in machine learning, visual explanation of ML processes, algorithmic explanation of ML models, human cognitive responses in ML-based decision making, human evaluation of machine learning and domain knowledge in transparent ML applications. This is the first book of its kind to systematically understand the current active research activities and outcomes related to human and machine learning. The book will not only inspire researchers to passionately develop new algorithms incorporating human for human-centred ML algorithms, resulting in the overall advancement of ML, but also help ML practitioners proactively use ML outputs for informative and trustworthy decision making. This book is intended for researchers and practitioners involved with machine learning and its applications. The book will especially benefit researchers in areas like artificial intelligence, decision support systems and human-computer interaction.

Making use of data is not anymore a niche project but central to almost every project. With access to massive compute resources and vast amounts of data, it seems at least in principle possible to solve any problem. However, successful data science projects result from the intelligent application of: human intuition in combination with computational power; sound background knowledge with computer-aided modelling; and critical reflection of the obtained insights and results. Substantially updating the previous edition, then entitled Guide to Intelligent Data Analysis, this core textbook continues to provide a hands-on instructional approach to many data science techniques, and explains how these are used to solve real world problems. The work balances the practical aspects of applying and using data science techniques with the theoretical and algorithmic underpinnings from mathematics and statistics. Major updates on techniques and subject coverage (including deep learning) are included. Topics and features: guides the reader through the process of data science, following the interdependent steps of project understanding, data understanding, data blending and transformation, modeling, as well as deployment and monitoring; includes numerous examples using the open source KNIME Analytics Platform, together with an introductory appendix; provides a review of the basics of classical statistics that support and justify many data analysis methods, and a glossary of statistical terms; integrates illustrations and case-study-style examples to support pedagogical exposition; supplies further tools and information at an associated website. This practical and systematic textbook/reference is a "need-to-have" tool for graduate and advanced undergraduate students and essential reading for all professionals who face data science problems. Moreover, it is a "need to use, need to keep" resource following one's exploration of the subject.

This handbook discusses challenges and limitations in existing solutions, and presents state-of-the-art advances from both academia and industry, in big data analytics and digital forensics. The second chapter comprehensively reviews IoT security, privacy, and forensics literature, focusing on IoT and unmanned aerial vehicles (UAVs). The authors propose a deep learning-based approach to process cloud's log data and mitigate enumeration attacks in the third chapter. The fourth chapter proposes a robust fuzzy learning model to protect IT-based infrastructure against advanced persistent threat (APT) campaigns. Advanced and fair clustering approach for industrial data, which is capable of training with huge volume of data in a close to linear time is introduced in the fifth chapter, as well as offering an adaptive deep learning model to detect cyberattacks targeting cyber physical systems (CPS) covered in the sixth chapter. The authors evaluate the performance of unsupervised machine learning for detecting cyberattacks against industrial control systems (ICS) in chapter 7, and the next chapter presents a robust fuzzy Bayesian approach for ICS's cyber threat hunting. This handbook also evaluates the performance of supervised machine learning methods in identifying cyberattacks against CPS. The performance of a scalable clustering algorithm for CPS's cyber threat hunting and the usefulness of machine learning algorithms for MacOS malware detection are respectively evaluated. This handbook continues with evaluating the performance of various machine learning techniques to detect the Internet of Things malware. The authors demonstrate how MacOSX cyberattacks can be detected using state-of-the-art machine learning models. In order to identify credit card frauds, the fifteenth chapter introduces a hybrid model. In the sixteenth chapter, the editors propose a model that leverages natural language processing techniques for generating a mapping between APT-related reports and cyber kill chain. A deep learning-based approach to detect ransomware is introduced, as well as a proposed clustering approach to detect IoT malware in the last two chapters. This handbook primarily targets professionals and scientists working in Big Data, Digital Forensics, Machine Learning, Cyber Security Cyber Threat Analytics and Cyber Threat Hunting as a reference book. Advanced level-students and researchers studying and working in Computer systems, Computer networks and Artificial intelligence will also find this reference useful.

Machine learning is currently one of the most rapidly growing areas of research in computer science. In compiling this volume we have brought together contributions from some of the most prestigious researchers in this field. This book covers the three main learning systems; symbolic learning, neural networks and genetic algorithms as well as providing a tutorial on learning casual influences. Each of the nine chapters is self-contained.

Both theoreticians and application scientists/engineers in the broad area of artificial intelligence will find this volume valuable. It also provides a useful sourcebook for Postgraduate since it shows the direction of current research.

Defines the notion of an activity model learned from sensor data and presents key algorithms that form the core of the field Activity Learning: Discovering, Recognizing and Predicting Human Behavior from Sensor Data provides an in-depth look at computational approaches to activity learning from sensor data. Each chapter is constructed to provide practical, step-by-step information on how to analyze and process sensor data. The book discusses techniques for activity learning that include the following: * Discovering activity patterns that emerge from behavior-based sensor data * Recognizing occurrences of predefined or discovered activities in real time * Predicting the occurrences of activities The techniques covered can be applied to numerous fields, including security, telecommunications, healthcare, smart grids, and home automation. An online companion site enables readers to experiment with the techniques described in the book, and to adapt or enhance the techniques for their own use. With an emphasis on computational approaches, Activity Learning: Discovering, Recognizing, and Predicting Human Behavior from Sensor Data provides graduate students and researchers with an algorithmic perspective to activity learning.

This book describes advanced machine learning models - such as temporal collaborative filtering, stochastic models and Bayesian nonparametrics - for analysing customer behaviour. It shows how they are used to track changes in customer behaviour, monitor the evolution of customer groups, and detect various factors, such as seasonal effects and preference drifts, that may influence customers' purchasing behaviour. In addition, the book presents four case studies conducted with data from a supermarket health program in which the customers were segmented and the impact of promotional activities on different segments was evaluated. The outcomes confirm that the models developed here can be used to effectively analyse dynamic behaviour and increase customer engagement. Importantly, the methods introduced here can also be used to analyse other types of behavioural data such as activities on social networks, and educational systems.

Artificial intelligence is at the forefront of research and implementation in many industries including healthcare and agriculture. Whether it's detecting disease or generating algorithms, deep learning techniques are advancing exponentially. Researchers and professionals need a platform in which they can keep up with machine learning trends and their developments in the real world. The Handbook of Research on Applications and Implementations of Machine Learning Techniques provides innovative insights into the multi-disciplinary applications of machine learning algorithms for data analytics. The content within this publication examines disease identification, neural networks, and language support. It is designed for IT professionals, developers, data analysts, technology specialists, R&D professionals, industrialists, practitioners, researchers, academicians, and students seeking research on deep learning procedures and their enactments in the fields of medicine, engineering, and computer science.

This book first presents a tutorial on Federated Learning (FL) and its role in enabling Edge Intelligence over wireless edge networks. This provides readers with a concise introduction to the challenges and state-of-the-art approaches towards implementing FL over the wireless edge network. Then, in consideration of resource heterogeneity at the network edge, the authors provide multifaceted solutions at the intersection of network economics, game theory, and machine learning towards improving the efficiency of resource allocation for FL over the wireless edge networks. A clear understanding of such issues and the presented theoretical studies will serve to guide practitioners and researchers in implementing resource-efficient FL systems and solving the open issues in FL respectively.

This book gathers selected papers presented at International Conference on Machine Learning, Advances in Computing, Renewable Energy and Communication (MARC 2020), held in Krishna Engineering College, Ghaziabad, India, during December 17-18, 2020. This book discusses key concepts, challenges, and potential solutions in connection with established and emerging topics in advanced computing, renewable energy, and network communications.