Networks of Learning Automata: Techniques for Online Stochastic Optimization is a comprehensive account of learning automata models with emphasis on multiautomata systems. It considers synthesis of complex learning structures from simple building blocks and uses stochastic algorithms for refining probabilities of selecting actions. Mathematical analysis of the behavior of games and feedforward networks is provided. Algorithms considered here can be used for online optimization of systems based on noisy measurements of performance index. Also, algorithms that assure convergence to the global optimum are presented. Parallel operation of automata systems for improving speed of convergence is described. The authors also include extensive discussion of how learning automata solutions can be constructed in a variety of applications.

This book collects recent theoretical advances and concrete applications of learning automata (LAs) in various areas of computer science, presenting a broad treatment of the computer science field in a survey style. Learning automata (LAs) have proven to be effective decision-making agents, especially within unknown stochastic environments. The book starts with a brief explanation of LAs and their baseline variations. It subsequently introduces readers to a number of recently developed, complex structures used to supplement LAs, and describes their steady-state behaviors. These complex structures have been developed because, by design, LAs are simple units used to perform simple tasks; their full potential can only be tapped when several interconnected LAs cooperate to produce a group synergy. In turn, the next part of the book highlights a range of LA-based applications in diverse computer science domains, from wireless sensor networks, to peer-to-peer networks, to complex social networks, and finally to Petri nets. The book accompanies the reader on a comprehensive journey, starting from basic concepts, continuing to recent theoretical findings, and ending in the applications of LAs in problems from numerous research domains. As such, the book offers a valuable resource for all computer engineers, scientists, and students, especially those whose work involves the reinforcement learning and artificial intelligence domains.

This book highlights both theoretical and applied advances in cellular learning automata (CLA), a type of hybrid computational model that has been successfully employed in various areas to solve complex problems and to model, learn, or simulate complicated patterns of behavior. Owing to CLA’s parallel and learning abilities, it has proven to be quite effective in uncertain, time-varying, decentralized, and distributed environments. The book begins with a brief introduction to various CLA models, before focusing on recently developed CLA variants. In turn, the research areas related to CLA are addressed as bibliometric network analysis perspectives. The next part of the book presents CLA-based solutions to several computer science problems in e.g. static optimization, dynamic optimization, wireless networks, mesh networks, and cloud computing. Given its scope, the book is well suited for all researchers in the fields of artificial intelligence and reinforcement learning.

This self-contained introductory text on the behavior of learning automata focuses on how a sequential decision-maker with a finite number of choices responds in a random environment. Topics include fixed structure automata, variable structure stochastic automata, convergence, 0 and S models, nonstationary environments, interconnected automata and games, and applications of learning automata. A must for all students of stochastic algorithms, this treatment is the work of two well-known scientists and is suitable for a one-semester graduate course in automata theory and stochastic algorithms. This volume also provides a fine guide for independent study and a reference for students and professionals in operations research, computer science, artificial intelligence, and robotics. The authors have provided a new preface for this edition.

This book begins by briefly explaining learning automata (LA) models and a recently developed cellular learning automaton (CLA) named wavefront CLA. Analyzing social networks is increasingly important, so as to identify behavioral patterns in interactions among individuals and in the networks’ evolution, and to develop the algorithms required for meaningful analysis. As an emerging artificial intelligence research area, learning automata (LA) has already had a significant impact in many areas of social networks. Here, the research areas related to learning and social networks are addressed from bibliometric and network analysis perspectives. In turn, the second part of the book highlights a range of LA-based applications addressing social network problems, from network sampling, community detection, link prediction, and trust management, to recommender systems and finally influence maximization. Given its scope, the book offers a valuable guide for all researchers whose work involves reinforcement learning, social networks and/or artificial intelligence.

This book is devoted to the leading research in applying learning automaton (LA) and heuristics for solving benchmark and real-world optimization problems. The ever-increasing application of the LA as a promising reinforcement learning technique in artificial intelligence makes it necessary to provide scholars, scientists, and engineers with a practical discussion on LA solutions for optimization. The book starts with a brief introduction to LA models for optimization. Afterward, the research areas related to LA and optimization are addressed as bibliometric network analysis. Then, LA's application in behavior control in evolutionary computation, and memetic models of object migration automata and cellular learning automata for solving NP hard problems are considered. Next, an overview of multi-population methods for DOPs, LA's application in dynamic optimization problems (DOPs), and the function evaluation management in evolutionary multi-population for DOPs are discussed. Highlighted benefits • Presents the latest advances in learning automata-based optimization approaches. • Addresses the memetic models of learning automata for solving NP-hard problems. • Discusses the application of learning automata for behavior control in evolutionary computation in detail. • Gives the fundamental principles and analyses of the different concepts associated with multi-population methods for dynamic optimization problems.

Emphasizing issues of computational efficiency, Michael Kearns and Umesh Vazirani introduce a number of central topics in computational learning theory for researchers and students in artificial intelligence, neural networks, theoretical computer science, and statistics. Emphasizing issues of computational efficiency, Michael Kearns and Umesh Vazirani introduce a number of central topics in computational learning theory for researchers and students in artificial intelligence, neural networks, theoretical computer science, and statistics. Computational learning theory is a new and rapidly expanding area of research that examines formal models of induction with the goals of discovering the common methods underlying efficient learning algorithms and identifying the computational impediments to learning. Each topic in the book has been chosen to elucidate a general principle, which is explored in a precise formal setting. Intuition has been emphasized in the presentation to make the material accessible to the nontheoretician while still providing precise arguments for the specialist. This balance is the result of new proofs of established theorems, and new presentations of the standard proofs. The topics covered include the motivation, definitions, and fundamental results, both positive and negative, for the widely studied L. G. Valiant model of Probably Approximately Correct Learning; Occam's Razor, which formalizes a relationship between learning and data compression; the Vapnik-Chervonenkis dimension; the equivalence of weak and strong learning; efficient learning in the presence of noise by the method of statistical queries; relationships between learning and cryptography, and the resulting computational limitations on efficient learning; reducibility between learning problems; and algorithms for learning finite automata from active experimentation.

Learning systems have made a significant impact on all areas of engineering problems. They are attractive methods for solving many problems which are too complex, highly non-linear, uncertain, incomplete or non-stationary, and have subtle and interactive exchanges with the environment where they operate. The main aim of the book is to give a systematic treatment of learning automata and to produce a guide to a wide variety of ideas and methods that can be used in learning systems, including enough theoretical material to enable the user of the relevant techniques and concepts to understand why and how they can be used. The book also contains the materials that are necessary for the understanding and development of learning automata for different purposes such as processes identification, optimization and control. Learning Automata: Theory and Applications may be recommended as a reference for courses on learning automata, modelling, control and optimization. The presentation is intended both for graduate students in control theory and statistics and for practising control engineers.

This book examines the intelligent random walk algorithms based on learning automata: these versions of random walk algorithms gradually obtain required information from the nature of the application to improve their efficiency. The book also describes the corresponding applications of this type of random walk algorithm, particularly as an efficient prediction model for large-scale networks such as peer-to-peer and social networks. The book opens new horizons for designing prediction models and problem-solving methods based on intelligent random walk algorithms, which are used for modeling and simulation in various types of networks, including computer, social and biological networks, and which may be employed a wide range of real-world applications.

Extensive data and mathematical theory are presented on the computer simulation or Probability State Variable (PSV) devices, e.g., the Neurotron, Random State Variable (RSV) Strategy devices, and networks of these elements as a function of the network connectivity, and score and value functions in four problem classes: re-entry trajectory prediction, pattern recognition, inferential measurement techniques, and self-organizing controllers. The simulation models are presented in detail. Hardware implementation is discussed, using statistical source design as a vehicle for insight into cost analysis of the Demonstrator Neurotron and the feasibility of construction of the networks studied. (Author).