Advances in Learning Automata and Intelligent Optimization (Intelligent Systems Reference Library, 208)

✍ Scribed by Javidan Kazemi Kordestani (editor), Mehdi Razapoor Mirsaleh (editor), Alireza Rezvanian (editor), Mohammad Reza Meybodi (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 355
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

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.

✦ Table of Contents

Preface
Contents
About the Authors
Abbreviations
1 An Introduction to Learning Automata and Optimization
1.1 Introduction
1.2 Learning Automata
1.2.1 Learning Automata Variants
1.2.2 Recent Applications of Learning Automata
1.3 Optimization
1.3.1 Evolutionary Algorithms and Swarm Intelligence
1.4 Reinforcement Learning and Optimization Methods
1.4.1 Static Optimization
1.4.2 Dynamic Optimization
1.5 LA and Optimization Timeline
1.6 Chapter Map
1.7 Conclusion
References
2 Learning Automaton and Its Variants for Optimization: A Bibliometric Analysis
2.1 Introduction
2.2 Learning Automata Models and Optimization
2.3 Material and Method
2.3.1 Data Collection and Initial Results
2.3.2 Refining the Initial Results
2.4 Analyzing the Results
2.4.1 Initial Result Statistics
2.4.2 Top Journals
2.4.3 Top Researchers
2.4.4 Top Papers
2.4.5 Top Affiliations
2.4.6 Top Keywords
2.5 Conclusion
References
3 Cellular Automata, Learning Automata, and Cellular Learning Automata for Optimization
3.1 Introduction
3.2 Preliminaries
3.2.1 Cellular Automata
3.2.2 Learning Automata
3.2.3 Cellular Learning Automata
3.3 CA, CLA, and LA Models for Optimization
3.3.1 Cellular Learning Automata-Based Evolutionary Computing (CLA-EC)
3.3.2 Cooperative Cellular Learning Automata-Based Evolutionary Computing (CLA-EC)
3.3.3 Recombinative Cellular Learning Automata-Based Evolutionary Computing (RCLA-EC)
3.3.4 CLA-EC with Extremal Optimization (CLA-EC-EO)
3.3.5 Cellular Learning Automata-Based Differential Evolution (CLA-DE)
3.3.6 Cellular Particle Swarm Optimization (Cellular PSO)
3.3.7 Firefly Algorithm Based on Cellular Learning Automata (CLA-FA)
3.3.8 Harmony Search Algorithm Based on Learning Automata (LAHS)
3.3.9 Learning Automata Based Butterfly Optimization Algorithm (LABOA)
3.3.10 Grey Wolf Optimizer Based on Learning Automata (GWO-LA)
3.3.11 Learning Automata Models with Multiple Reinforcements (MLA)
3.3.12 Cellular Learning Automata Models with Multiple Reinforcements (MCLA)
3.3.13 Multi-reinforcement CLA with the Maximum Expected Rewards (MCLA)
3.3.14 Gravitational Search Algorithm Based on Learning Automata (GSA-LA)
3.4 Conclusion
References
4 Learning Automata for Behavior Control in Evolutionary Computation
4.1 Introduction
4.2 Types of Parameter Adjustment in EC Community
4.2.1 EC with Constant Parameters
4.2.2 EC with Time-Varying Parameters
4.3 Differential Evolution
4.3.1 Initialization
4.3.2 Difference-Vector Based Mutation
4.3.3 Repair Operator
4.3.4 Crossover
4.3.5 Selection
4.4 Learning Automata for Adaptive Control of Behavior in Differential Evolution
4.4.1 Behavior Control in DE with Variable-Structure Learning Automaton
4.4.2 Behavior Control in DE with Fixed-Structure Learning Automaton
4.5 Experimental Setup
4.5.1 Benchmark Functions
4.5.2 Algorithm’s Configuration
4.5.3 Simulation Settings and Results
4.5.4 Experimental Results
4.6 Conclusion
References
5 A Memetic Model Based on Fixed Structure Learning Automata for Solving NP-Hard Problems
5.1 Introduction
5.2 Fixed Structure Learning Automata and Object Migrating Automata
5.2.1 Fixed Structure Learning Automata
5.2.2 Object Migration Automata
5.3 GALA
5.3.1 Global Search in GALA
5.3.2 Crossover Operator
5.3.3 Mutation Operator
5.3.4 Local Learning in GALA
5.3.5 Applications of GALA
5.4 The New Memetic Model Based on Fixed Structure Learning Automata
5.4.1 Hybrid Fitness Function
5.4.2 Mutation Operators
5.4.3 Crossover Operators
5.5 The OneMax Problem
5.5.1 Local Search for OneMax
5.5.2 Experimental Results
5.6 Conclusion
References
6 The Applications of Object Migration Automaton (OMA)-Memetic Algorithm for Solving NP-Hard Problems
6.1 Introduction
6.2 The Equipartitioning Problem
6.2.1 Local Search for EPP
6.2.2 Experimental Results
6.3 The Graph Isomorphism Problem
6.3.1 The Local Search in the Graph Isomorphism Problem
6.3.2 Experimental Results
6.4 Assignment of Cells to Switches Problem (ACTSP) in Cellular Mobile Network
6.4.1 Background and Related Work
6.4.2 The OMA-MA for Assignment of Cells to Switches Problem
6.4.3 The Framework of the OMA-MA Algorithm
6.4.4 Experimental Result
6.5 Conclusion
References
7 An Overview of Multi-population Methods for Dynamic Environments
7.1 Introduction
7.2 Moving Peaks Benchmark
7.2.1 Extended Versions of MPB
7.3 Performance Measurement
7.4 Types of Multi-population Methods
7.4.1 Methods with a Fixed Number of Populations
7.4.2 Methods with a Variable Number of Populations
7.4.3 Methods Based on Population Clustering
7.4.4 Self-adapting the Number of Populations
7.5 Numerical Results
7.6 Conclusions
References
8 Learning Automata for Online Function Evaluation Management in Evolutionary Multi-population Methods for Dynamic Optimization Problems
8.1 Introduction
8.2 Preliminaries
8.2.1 Waste of FEs Due to Change Detection
8.2.2 Waste of FEs Due to the Excessive Number of Sub-populations
8.2.3 Waste of FEs Due to Overcrowding of Subpopulations in the Same Area of the Search Space
8.2.4 Waste of FEs Due to Exclusion Operator
8.2.5 Allocation of FEs to Unproductive Populations
8.2.6 Unsuitable Parameter Configuration of the EC Methods
8.2.7 Equal Distribution of FEs Among Sub-populations
8.3 Theory of Learning Automata
8.3.1 Fixed Structure Learning Automata
8.3.2 Variable Structure Learning Automata
8.4 EC Techniques under Study
8.4.1 Particle Swarm Optimization
8.4.2 Firefly Algorithm
8.4.3 Jaya
8.5 LA-Based FE Management Model for MP Evolutionary Dynamic Optimization
8.5.1 Initialization of Sub-populations
8.5.2 Detection and Response to Environmental Changes
8.5.3 Choose a Sub-population for Execution
8.5.4 Evaluate the Search Progress of Populations and Generate the Reinforcement Signal
8.5.5 Exclusion
8.6 FE-Management in MP Method with a Fixed Number of Populations
8.6.1 VSLA-Based FE Management Strategy
8.6.2 FSLA-Based FE Management Strategies
8.7 Experimental Study
8.7.1 Experimental Setup
8.7.2 Experimental Results and Discussion
8.8 Conclusion
References
9 Function Management in Multi-population Methods with a Variable Number of Populations: A Variable Action Learning Automaton Approach
9.1 Introduction
9.2 Main Framework of Clustering Particle Swarm Optimization
9.2.1 Creating Multiple Sub-swarms from the Cradle Swarm
9.2.2 Local Search by PSO
9.2.3 Status of Sub-swarms
9.2.4 Detection and Response to Environmental Changes
9.3 Variable Action-Set Learning Automata
9.4 FEM in MP Methods with a Variable Number of Populations
9.5 Experimental Study
9.5.1 Dynamic Test Function
9.5.2 Performance Measure
9.5.3 Experimental Settings
9.5.4 Experimental Results
9.6 Conclusions
References