Learning Automata and Their Applications to Intelligent Systems

About the Authors Preface and Acknowledgements A Guide to Reading This Book 1 Introduction 1.1 Ranking and selection in Noisy Optimization 1.2 Learning Automata and Ordinal Optimization 1.3 Exercises References 2 Learning Automata 2.1 Environment and Automaton 2.1.1 The Environment 2.1.2 The Automaton 2.1.3 Deterministic and Stochastic Automata 2.1.4 Measured Norms 2.2 Fixed Structure Learning Automata 2.2.1 Tsetlin Learning Automaton 2.2.2 Krinsky Learning Automaton 2.2.3 Krylov Learning Automaton 2.2.4 IJA Learning Automaton 2.3 Variable Structure Learning Automata 2.3.1 Estimator-Free Learning Automaton 2.3.2 Deterministic Estimator Learning Automaton 2.3.3 Stochastic Estimator Learning Automaton 2.4 Summary 2.5 Exercises References 3 Fast Learning Automata 3.1 Last-position Elimination-based Learning Automata 3.1.1 Background and Motivation 3.1.2 Principles and Algorithm Design 3.1.3 Difference Analysis 3.1.4 Simulation Studies 3.1.5 Summary 3.2 Fast Discretized Pursuit Learning Automata 3.2.1 Background and Motivation 3.2.2 Algorithm Design of Fast Discretized Pursuit LA 3.2.3 Optimality Analysis 3.2.4 Simulation Studies 3.2.5 Summary 3.3 Exercises References 4 Application-Oriented Learning Automata (55 pages) 4.1 Discovering and Tracking Spatiotemporal Event Patterns 4.1.1 Background and Motivation 4.1.2 Spatiotemporal Pattern Learning Automata 4.1.3 Adaptive Tunable Spatiotemporal Pattern Learning Automata 4.1.4 Optimality Analysis 4.1.5 Simulation Studies 4.1.6 Summary 4.2 Stochastic Searching on the Line 4.2.1 Background and Motivation 4.2.2 Symmetrical Hierarchical Stochastic Searching on the Line 4.2.3 Simulation Studies 4.2.4 Summary 4.3 Fast Adaptive Search on the Line in Dual Environments 4.3.1 Background and Motivation 4.3.2 Symmetrized ASS with Buffer 4.3.3 Simulation Studies 4.3.4 Summary 4.4 Exercises References 5 Ordinal Optimization (51 pages) 5.1 Optimal Computing-Budget Allocation 5.2 Optimal Computing-Budget Allocation for Selection of Best and Worst Designs 5.2.1 Background and Motivation 5.2.2 Approximate Optimal Simulation Budget Allocation 5.2.3 Simulation Studies 5.2.4 Summary 5.3 Optimal Computing-Budget Allocation for Subset Ranking 5.3.1 Background and Motivation 5.3.2 Approximate Optimal Simulation Budget Allocation 5.3.3 Simulation Studies 5.3.4 Summary 5.4 Exercises References 6 Incorporation of Ordinal Optimization into Learning Automata (22 pages) 6.1 Background and Motivation 6.2 Learning Automata with Optimal Computing Budget Allocation 6.3 Proof of Optimality 6.4 Simulation Studies 6.5 Summary 6.6 Exercises References 7 Noisy Optimization Applications (31 pages) 7.1 Background and Motivation 7.2 Particle Swarm Optimization 7.2.1 Parameters Configurations 7.2.2 Topology Structures 7.2.3 Hybrid PSO 7.2.4 Multiswarm Techniques 7.3 Resampling for Noisy Optimization Problems 7.4 PSO-Based LA and OCBA 7.5 Simulation Studies 7.6 Summary 7.7 Exercises References 8 Applications and Future Research Directions of Learning Automata (19 pages) 8.1 Summary of Existing Applications 8.1.1 Classification 8.1.2 Clustering 8.1.3 Games 8.1.4 Knapsack Problems 8.1.5 Decision Problems in Networks 8.1.6 Optimization 8.1.7 LA Parallelization and Design Ranking 8.1.8 Scheduling 8.2 Future Research Directions 8.3 Exercises References Index