Hybrid Quantum Metaheuristics: Theory and Applications

Description

The reference text introduces the principles of quantum mechanics to evolve hybrid metaheuristics-based optimization techniques useful for real world engineering and scientific problems. The text covers advances and trends in methodological approaches, theoretical studies, mathematical and applied techniques related to hybrid quantum metaheuristics and their applications to engineering problems. The book will be accompanied by additional resources including video demonstration for each chapter. It will be a useful text for graduate students and professional in the field of electrical engineering, electronics and communications engineering, and computer science engineering, this text: Discusses quantum mechanical principles in detail. Emphasizes the recent and upcoming hybrid quantum metaheuristics in a comprehensive manner. Provides comparative statistical test analysis with conventional hybrid metaheuristics. Highlights real-life case studies, applications, and video demonstrations. Cover Half Title Series Page Title Page Copyright Page Dedication Contents Editors Preface Contributors Chapter 1: An Introductory Illustration to Quantum-Inspired Metaheuristics 1.1. Introduction 1.2. Quantum-Inspired Metaheuristics 1.2.1. Local Search Metaheuristics 1.2.2. Constructive Metaheuristics 1.2.3. Population-based Metaheuristics 1.2.4. Hybrid Metaheuristics 1.3. Entanglement-Induced Optimization 1.4. W-state Encoding of Optimization Algorithms 1.5. Quantum System-based Optimization 1.5.1. Bi-level Quantum System-based Optimization 1.5.2. Multi-level Quantum System-based Optimization 1.6. Applications of Quantum-Inspired Metaheuristics 1.7. Conclusion Chapter 2: A Quantum-Inspired Approach to Collective Combine Basic Classifiers 2.1. Introduction 2.2. Bagging Method 2.3. Classifiers Based on Similarity of Objects 2.4. Statistical Classification Algorithms 2.5. Classifiers Based on Class Separability in Attribute Space 2.6. Logical Classification Algorithms 2.7. Neural Networks 2.8. Methods of Combining Basic Classifiers 2.8.1. Voting 2.8.2. Stacking 2.8.3. Ensemble Selection 2.8.3.1. The Bayesian Classifier 2.8.4. Quantum-Inspired Metaheuristics Method 2.9. Conclusion Chapter 3: Function Optimization Using IBM Q 3.1. Introduction 3.2. Function Optimization 3.2.1. Difficulties in Optimization Methods 3.2.2. Definition of Multi-objective Optimization Problem (MOOP) 3.2.3. Differences between SOOPs and MOOPs 3.3. Modern Optimization Problem-Solving Techniques 3.3.1. Genetic Algorithm 3.3.2. Simulated Annealing 3.3.3. Particle Swarm Optimization 3.3.4. Bat Algorithm 3.3.5. Cuckoo Search Algorithm 3.3.6. Fuzzy System 3.3.7. Neural Network Based Optimization 3.4. Quantum Computing and Optimization Algorithms 3.4.1. Quantum Computing 3.4.2. Optimization Using Quantum Computing 3.5. Features of IBM Q Experience 3.6. Circuit Composer IBM Q 3.7. QISKit in IBM Q 3.7.1. Creating 5-qubit Circuit with the Help of QISKit in IBM Q 3.7.2. Testing the Circuit Using IBM Quantum Computer 3.8. Optimization Using IBM Q 3.9. Conclusion Chapter 4: Multipartite Adaptive Quantum-Inspired Evolutionary Algorithm to Reduce Power Losses 4.1. Introduction 4.2. Literature Review 4.3. Problem Formulation 4.4. Power Flow 4.5. Algorithm 4.6. Results and Discussion 4.7. Conclusions 4.8. Parameters of IEEE Benchmark Test Bus System Chapter 5: Quantum-Inspired Manta Ray Foraging Optimization Algorithm for Automatic Clustering of Color Images 5.1. Introduction 5.2. Literature Review 5.3. Fundamentals of Quantum Computing 5.3.1. Rotation Gate 5.3.2. Pauli-X Gate 5.4. Validity Measurement of Clustering 5.5. Overview of Manta Ray Foraging Optimization Algorithm 5.6. Proposed Methodology 5.7. Experimental Results and Analysis 5.7.1. Developmental Entertainment 5.7.2. Dataset Used 5.7.3. Clustered Images 5.7.4. Sensitivity Analysis of QIMRFO 5.7.5. Analysis of Experimental Results 5.8. Conclusion and Future Scope Chapter 6: Automatic Feature Selection for Coronary Stenosis Detection in X-Ray Angiograms 6.1. Introduction 6.2. Background 6.2.1. Feature Extraction 6.2.1.1. Pixel Intensity-based Features 6.2.1.2. Texture Features 6.2.1.3. Morphologic Features 6.2.2. Feature Selection 6.2.3. Support Vector Machines 6.2.4. Quantum Genetic Algorithm 6.3. Proposed Method 6.4. Experiment Details 6.5. Results 6.6. Conclusion Chapter 7: Quantum Preprocessing for DCNN in Atherosclerosis Detection 7.1. Introduction 7.2. Related Work 7.3. Mathematical Foundations 7.3.1. Quantum Computing 7.3.1.1. Qubit States 7.3.1.2. Qubit Operations 7.3.1.3. Qubit Measurements 7.3.2. Convolutional Neural Networks 7.3.2.1. Convolutional Layer 7.3.2.2. Pooling Layer 7.3.2.3. Fully Connected Layer 7.3.2.4. Activation Functions 7.4. Proposed Method 7.4.1. Quantum Convolutional Layer 7.4.2. Network Architecture 7.4.3. Evaluation Metrics 7.5. Results and Discussions 7.5.1. Dataset of Coronary Stenosis 7.5.2. Quantum Preprocessing 7.5.3. Training Results 7.5.4. Detection Results 7.6. Concluding Remarks Chapter 8: Multilevel Quantum Elephant Herd Algorithm for Automatic Clustering of Hyperspectral Images 8.1. Introduction 8.2. Literature Survey 8.3. Background Concepts 8.3.1. Elephant Herding Optimization 8.3.1.1. Clan Updation 8.3.1.2. Separation Operator 8.3.1.3. Steps of EHO 8.3.2. Basic Concepts of Quantum Computing 8.3.3. Fuzzy C Means Clustering Algorithm 8.3.4. Xie-Beni Index 8.4. Proposed Methodology 8.4.1. HSI Preprocessing 8.4.2. Qubit and Qutrit Based Elephant Herd Optimization 8.5. Experimental Results and Analysis 8.5.1. Salinas Dataset 8.5.2. Fitness Function 8.5.3. Analysis 8.6. Conlusion Chapter 9: Toward Quantum-Inspired SSA for Solving Multiobjective Optimization Problems 9.1. Introduction 9.2. Salp Swarm Algorithm 9.2.1. Initialization 9.2.2. Leaders’ Specification 9.2.3. Updating Position 9.2.4. Re-evaluation and Decision-making 9.3. Proposed Multiobjective Quantum-inspired Salp Swarm Algorithm 9.3.1. Delta Potential-well Model for SSA 9.3.2. Salp Position Measurement 9.3.3. The New Algorithm Behavior 9.4. Experimental Procedure 9.4.1. Computing Environment 9.4.2. Performance Assessment Metrics 9.4.3. Multiobjective Benchmark Problems 9.4.4. Evaluating Method and Algorithms Parameters 9.5. Experiments and Discussion 9.6. Conclusion Chapter 10: Quantum-Inspired Multi-Objective NSGA-II Algorithm for Automatic Clustering of Gray Scale Images 10.1. Introduction 10.2. Quantum Computing Fundamental 10.3. Computing the Objectives 10.3.1. CS-Measure (CSM) index 10.3.2. Davies–Bouldin (DB) Index 10.4. Multi-Objective Optimization 10.4.1. NSGA-II 10.4.2. Population Initialization and Chromosome Representation 10.4.3. Creating Cluster Centroids 10.4.4. Genetic Operation 10.4.5. Fast Non-dominated Sorting 10.4.6. Crowding Distance 10.4.7. Basic Steps of Classical NSGA-II Algorithm for Automatic Clustering of Gray Scale Images 10.5. Proposed Technique 10.5.1. Quantum State Population Initialization 10.5.2. Creating Cluster Centroids in Quantum-Inspired Framework 10.5.3. Genetic Operators in Quantum-Inspired Framework 10.5.3.1. Quantum-Behaved Selection 10.5.3.2. Quantum-Behaved Crossover 10.5.3.3. Quantum-Behaved Mutation 10.5.4. Fast Non-dominated Sorting in Quantum-Inspired Framework 10.5.5. Crowding Distance Computation in Quantum-Inspired Framework 10.5.6. QIMONSGA-II Algorithm for Automatic Clustering of Gray Scale Images 10.6. Experimental Results and Analysis 10.6.1. Used Dataset 10.6.2. Parameter Settings 10.6.3. Performance Evaluation 10.6.4. Experimental Results 10.7. Discussions and Conclusion Chapter 11: Conclusion Appendix A: Automatic Feature Selection for Coronary Stenosis Detection in X-Ray Angiograms A.1. Matlab Code to Extract Vessel Segments A.2. Matlab Code to Find Pixel Positions A.3. Matlab Code to Extract a Window from a Matrix A.4. Matlab Code to Find Row Vector Index