Advances in Swarm Intelligence : 14th International Conference, ICSI 2023, Shenzhen, China, July 14–18, 2023, Proceedings, Part II

This two-volume set LNCS 13968 and 13969 constitutes the proceedings of the 14 th International Conference on Advances in Swarm Intelligence, ICSI 2023, which took place in Shenzhen, China, China, in July 2023. The theme of this year’s conference was “Serving Life with Swarm Intelligence”. The 81 full papers presented were carefully reviewed and selected from 170 submissions. The papers are organized into 12 cohesive sections covering major topics of swarm intelligence research and its development and applications. The papers of the second part cover topics such as: Swarm Robotics and UAV; Machine Learning; Data Mining; Routing and Scheduling Problems; Stock Prediction and Portfolio Optimization; ICSI-Optimization Competition. Preface Organization Contents – Part II Contents – Part I Swarm Robotics and UAV A Blockchain-Based Service-Oriented Framework to Enable Cooperation of Swarm Robots 1 Introduction 2 A Blockchain-Based Framework for Swarm Robots 2.1 Framework 2.2 Service Registration and State Update 2.3 Service Search and Task Tracking 2.4 Service Composition 2.5 Service Trading 3 Experiments and Results 3.1 Experimental Setup 3.2 Evaluation of Service Planning 3.3 Evaluation of Service Trading 4 Conclusion References Collective Behavior for Swarm Robots with Distributed Learning 1 Introduction 2 Flocking Control Algorithm 2.1 Collective Behavior 2.2 Models for Mobile Robots 2.3 Distributed Learning 2.4 Cooperative Control 3 Convergence Analysis 4 Simulation Results 4.1 Standard Environment 4.2 Noise Environment 5 Conclusion References Exploration of Underwater Environments with a Swarm of Heterogeneous Surface Robots 1 Introduction 2 Leader-Follower Collective Motion 2.1 Swarm Collective Motion of Followers 2.2 Controller of the Leader 3 Realisation of the Exploration 3.1 Robotic Platforms 3.2 Simulation Platform 3.3 Experiments 4 Results and Discussion 4.1 Position Error of the Leader 4.2 Performance of Swarm in Simple Trajectory 4.3 Performance of Swarm in Exploration 4.4 The Position Error of Followers 4.5 Summary 5 Conclusion References Integrating Reinforcement Learning and Optimization Task: Evaluating an Agent to Dynamically Select PSO Communication Topology*-1pc 1 Introduction 2 Particle Swarm Optimization 3 Methodology 4 Results 5 Conclusions References Swarm Multi-agent Trapping Multi-target Control with Obstacle Avoidance 1 Introduction 2 Modeling 3 Details of Trapping Control with Obstacle Avoidance 4 Simulation Results and Discussion 4.1 Simulation Results 4.2 Discussion 5 Conclusion References A Novel Data Association Method for Multi-target Tracking Based on IACA 1 Introduction 2 Combinatorial Optimization Model of Data Association 3 Improved Ant Colony Algorithm and Its Application in Data Association 3.1 Research on the Traditional Ant Colony Algorithm 3.2 Our Improvement Strategies in Ant Colony Algorithm 3.3 IACDA Method 4 Simulation Experiment and Analysis 5 Conclusion References MACT: Multi-agent Collision Avoidance with Continuous Transition Reinforcement Learning via Mixup*-1pc 1 Introduction 2 Preliminaries 2.1 Problem Formulation 2.2 Mixup 3 Methodologies 3.1 Reinforcement Learning Setup 3.2 Multi-agent Continuous Transition 4 Experiments 4.1 Simulation 4.2 Real World 5 Conclusion References Research on UAV Dynamic Target Tracking with Multi-sensor Position Feedback*-12pt 1 Instruction 2 Path Planning Algorithm Design 3 Design of Control Law Based on Backstepping Method 4 UWB and INS Fusion Location Algorithm Based on UKF 5 Simulation and Analysis 6 Conclusion References Multiple Unmanned Aerial Vehicles Path Planning Based on Collaborative Differential Evolution 1 Introduction 2 Mathematical Model of Multi-UAVs Path Planning 2.1 Path Representation 2.2 Cost and Constraint 3 Path Planning Based on CODE 3.1 Standard Differential Evolution 3.2 Collaborative Differential Evolution 3.3 Path Planning for Multi-UAVs 4 Experiment Evaluation and Comparison 5 Conclusion References Design and Analysis of VLC-OCC-CDMA Rake System 1 Introduction 2 System Model 2.1 Transmitter 2.2 Channel Model 3 Design and Analysis for VLC-OCC-CDMA Rake Receiver 3.1 Design of Rake Receiver for VLC-OCC-CDMA System 3.2 Analysis of Rake Receiver for VLC-OCC-CDMA System 3.3 Theoretical Analysis on Elimination of Interference and BER 4 Simulation Details and Results 4.1 Simulation Setup 4.2 Simulation Results 5 Conclusions References Machine Learning Noise-Tolerant Hardware-Aware Pruning for Deep Neural Networks 1 Introduction 2 Related Work 3 Methodology 3.1 Pruning as Optimization 3.2 Noise in Hardware-aware Pruning 3.3 Optimization Algorithm 4 Experiments 4.1 Experimental Setup 4.2 Effectiveness of Noise-Tolerant Hardware-Aware Pruning 4.3 Comparison of Pruning Performances 5 Conclusion References Pyrorank: A Novel Nature-Inspired Algorithm to Promote Diversity in Recommender Systems 1 Introduction 2 Related Work 2.1 Related Work: Recommendation Diversity Metrics 2.2 Related Work: Diversification Algorithms 2.3 Biological Inspired Approach: Pyrodiversity 3 Pyrorank: Algorithm Design and Implementation 3.1 Pyrorank: High Level Design 3.2 Pyrorank Algorithm: Implementation 3.3 Pyrorank: Pseudo-code 3.4 Pyrorank: Runtime Analysis 4 Experiments and Results 4.1 Recommendation Model 4.2 Datasets 4.3 Feature Engineering 4.4 Diversity Measures 4.5 Algorithms 4.6 Comparisons 5 Discussion 6 Conclusion 6.1 Future Work and Research References Analysis of SIR Compartmental Model Results with Different Update Strategies 1 Introduction 2 Overview of the SIR Model 2.1 Universal Aspects of Models 2.2 Specifics of the SIR Model 3 Theoretical Setup 3.1 Approaches to Iterating 4 Experimental Setup and Results 4.1 Dynamics of the Model 4.2 Graphical Results 4.3 Numerical Results 5 Conclusion References Research on Location Selection of General Merchandise Store Based on Machine Learning 1 Introduction 2 Feature Engineering 2.1 Original Dataset 2.2 Feature Engineering 3 Location Selection of General Merchandise Store Based on Machine Learning 3.1 Dataset Construction 3.2 The Single Models of Location Selection of General Merchandise Store 3.3 Stacking Model of Location Selection of General Merchandise Store 4 The Application of the Location Selection Model 5 Conclusion Appendix References CF-PMSS: Collaborative Filtering Based on Preference Model and Sparrow Search 1 Introduction 2 Problem Description 3 Preference Model 4 SSA-Enhanced K-Means Clustering 4.1 Basic Principle of SSA 4.2 Determining Cluster Centers with SSA 5 Synthesis Similarity Measure 5.1 Type-Based Similarity 5.2 Combining Rating and Type 6 Overall Algorithm 7 Performance Evaluation 7.1 Evaluation Metrics 7.2 Parameter Settings 7.3 Comparison Results 8 Conclusions References Asynchronous Federated Learning Framework Based on Dynamic Selective Transmission*-1pc 1 Introduction 2 Related Work 2.1 ASTW 2.2 DESC 3 The Proposed Framework 3.1 AFL-DST 3.2 Dynamic Selective Transmission Strategy 4 Experiments 5 Conclusion References Data Mining Small Aerial Target Detection Algorithm Based on Improved YOLOv5 1 Introduction 2 Related Work 2.1 Small Object Detection 2.2 YOLOv5 3 Methodology 3.1 Optimize the Scale of Prediction 3.2 Fusing BiFP 4 Experiments 4.1 Dataset 4.2 Evaluation Indicators 4.3 Experimental Environment 4.4 The Results of the Experiment 5 Conclusion References Secondary Pulmonary Tuberculosis Lesions Detection Based on Improved YOLOv5 Networks*-1pc 1 Introduction 2 Related Research 3 Methods and Materials 3.1 Dataset Description 3.2 Data Augmentation 3.3 Improved YOLOv5 Network 4 Experimental Results 4.1 Evaluation Metrics 4.2 Model Training 4.3 Comparisons of Various Neural Networks 4.4 Performance Analysis on Cross-Hospital Dataset 5 Conclusion References Abnormal Traffic Detection Based on a Fusion BiGRU Neural Network 1 Introduction 2 Related Work 2.1 Abnormal Traffic Detection 2.2 Deep Learning 3 Method 3.1 Data Pre-processing 3.2 Spatial Feature Extraction 3.3 Timing Feature Extraction 3.4 Soft-Max Classifier 4 Experiment 4.1 Dataset 4.2 Evaluation Metrics and Model Training 5 Experimental Results and Analysis 6 Ablation Experiments 6.1 Dual Branch Comparison 6.2 Hyperparameter Epoch 7 Conclusion References A Fabric Defect Detection Model Based on Feature Extraction of Weak Sample Scene 1 Introduction 2 Related Work 2.1 Traditional Cloth Defect Detection 2.2 Fabric Defect Detection Based on Deep Learning 3 Methods 3.1 Structure 3.2 Defective Feature Extraction Backbone 3.3 Anomaly Score Calculation 3.4 Acceleration 4 Experiments 4.1 Accuracy 4.2 Time Elapse 5 Conclusion References Intrusion Detection Method Based on Complementary Adversarial Generation Network 1 Introduction 2 Related Work 2.1 Intrusion Detection 2.2 Generative Adversarial Network 3 Method 3.1 Complementary Confrontation Generates Adversarial Networks 3.2 Stacked Non-symmetry Autoencoder 4 Experiment 4.1 Environment 4.2 Process of Experiment 5 Conclusion References EEG-Based Subject-Independent Depression Detection Using Dynamic Convolution and Feature Adaptation 1 Introduction 2 Methods 2.1 Overview 2.2 Dynamic Feature Extractor 2.3 Domain Adaptation Scheme 3 Experiment 3.1 Datasets 3.2 State-of-the-Art Models 3.3 Experimental Settings 3.4 Evaluation Metrics 4 Results 5 Conclusions References Multi-label Adversarial Defense Scheme Based on Negative Correlation Ensemble 1 Introduction 2 Related Work 2.1 Ensemble Defense 2.2 Negative Correlation 3 The Proposed Method 3.1 Gradient Direction 3.2 Gradient Magnitude 3.3 ML-NCEn 4 Experiments 4.1 Experimental Setup 4.2 Experimental Results and Analysis 5 Conclusion References Analysis of the Impact of Mathematics Courses on Professional Courses in Science and Engineering Majors 1 Introduction 2 The Theoretical Basis of the Algorithm Used 2.1 Pearson Correlation Coefficient [18] 2.2 Canonical Correlation Analysis 3 Data Processing and Analysis of Results 3.1 Statistical Descriptive Analysis of Course Grades 3.2 Correlation Coefficient Between Courses 3.3 Canonical Correlation Analysis Between Course Classes 4 Conclusion and Awareness References Intelligent System Reliability Modeling Methods 1 Introduction 2 Failure Mechanism, Risk and Applicability Analysis of Traditional Modeling Methods 3 Intelligent System Fault Coupling Relationship Analysis Method 3.1 Overview of Causal Analysis Methods 3.2 Quantitative Analysis of Fault Coupling Relationship Based on Causal Analysis 3.3 Reliability Modeling Method Considering Fault Coupling Relationship 4 Case Study 5 Conclusion and Foresight References Sensitivity of Clustering Algorithms to Sparseness of One Correlation Network 1 Introduction 2 Theoretical Background 2.1 Correlation Network 2.2 Graph Clustering Algorithms 2.3 Cluster Validity Indices 3 Research 3.1 Data Processing Software 3.2 Data Description 3.3 Results 4 Conclusion References Routing and Scheduling Problems Monte Carlo Tree Search with Adaptive Estimation for DAG Scheduling 1 Introduction 2 DAG Scheduling Problem Description 2.1 Mathematical Problem Statement 3 Algorithm 3.1 Algorithm Architecture 3.2 Graph Scheduling Metrics 3.3 State-of-the-Art DAG Scheduling Algorithms 4 Numerical Simulation 4.1 Simulation Environment 4.2 Training and Testing Graph Description 4.3 Training of Neural Network 4.4 Result Compared with SoTA 4.5 Detailed Comparison 5 Conclusion References Resource Allocation in Heterogeneous Network with Supervised GNNs 1 Introduction 2 System Model 3 Graph Neural Networks Based Approach 3.1 Dataset Generation 3.2 Graph Representation 3.3 Graph Neural Network 3.4 Unsupervised Learning 3.5 Supervised Learning 4 Simulation Results and Analysis 4.1 Simulation Setup 4.2 Performance Comparison 5 Conclusion References Satellite Downlink Scheduling Under Breakpoint Resume Mode 1 Introduction 2 Problem Analysis 2.1 The Input and Output of the SDSP 2.2 Assumptions 2.3 Constraints 2.4 Problem Complexity Analysis 2.5 Mixed Integer Programming Model for SDSP-BRM 3 A Simple and Effective Heuristic Algorithm (SEHA) for SDSP-BRM 3.1 Solution Construction 3.2 Operators 3.3 Termination Criterions 4 Experimental Study 4.1 Experimental Setup 4.2 Comparative Results of SEHA with CPLEX 4.3 The Impact of the Segmental Strategy 5 Conclusions References A Repetitive Grouping Max-Min Ant System for Multi-Depot Vehicle Routing Problem with Time Window 1 Introduction 2 Problem Description 3 The Proposed Algorithm RG-MMAS 3.1 Grouping Phase 3.2 Adaptive Range 3.3 Local Search 4 Experimental Results 5 Conclusion References Secure Access Method of Power Internet of Things Based on Zero Trust Architecture 1 Introduction 2 Network Security Risks Faced by Electric Internet of Things 2.1 The Security Boundary of Electric Internet of Things is Fuzzy 2.2 Electric Iot Faces Internal and External Attack Risks 3 Zero-Trust Architecture 3.1 The Proposal and Development of Zero Trust Concept 3.2 Zero-Trust Architecture 3.3 Zero-Trust Architecture Solution 4 Security Framework of Electric Internet of Things Based on Zero Trust 4.1 Design of Security Protection Framework 4.2 Security Authentication 4.3 Dynamic Access Control 5 Application Scenario Validation 5.1 Distributed Power Access Scenarios 5.2 Analysis of the Attack of Distributed Power Terminals 6 Conclusion References On the Complete Area Coverage Problem of Painting Robots 1 Introduction 2 Background 3 Algorithm and Its Illustration 3.1 Algorithm 3.2 Illustration of Algorithm 3.3 Illustration of Algorithm 4 Performance Analysis 5 Experimental Results 6 Concluding Remarks and Future Works References Reachability Map-Based Motion Planning for Robotic Excavation 1 Introduction 2 Related Works 2.1 Motion Planning Methods 2.2 Reachability Map 3 Proposed Method 3.1 Overview 3.2 Smoothness 3.3 Reachablility Map 3.4 Graph Construction 3.5 Cost Function 3.6 Heuristic 4 Experiment 4.1 Implementation Details 4.2 Motion Planning in Free Space 4.3 Motion Planning During Trenching 5 Conclusions References Reinforced Vision-and-Language Navigation Based on Historical BERT*-1pc 1 Introduction 2 Vision-and-Language Navigation Based on Pre-training 2.1 Pre-trained Multi-modal BERT Model for VLN 2.2 Recurrent BERT Model for Vision-and-Language Navigation 2.3 Training with Reinforcement Learning and Imitation Learning 3 Experiments 3.1 Dataset and Simulator for Indoor Scene Vision-and-Language Navigation 3.2 Evaluation Metrics 3.3 Experiment Analysis 3.4 Ablation Study 4 Conclusion References Stock Prediction and Portfolio Optimization Meta–heuristics for Portfolio Optimization: Part I — Review of Meta–heuristics 1 Introduction 2 Mean–variance Portfolio Optimization 3 Meta–heuristics for Portfolio Optimization 3.1 Artificial Bee Colony 3.2 Firefly Algorithm 3.3 Genetic Algorithm 3.4 Particle Swarm Optimization 3.5 Set–based Particle Swarm Optimization 4 Conclusion A Set Operators References Meta-heuristics for Portfolio Optimization: Part II—Empirical Analysis 1 Introduction 2 Empirical Process 2.1 Benchmark Problems 2.2 Control Parameter Tuning 2.3 Performance Measures 3 Results 3.1 Hang Seng 3.2 DAX 100 3.3 FTSE 100 3.4 S and P 100 3.5 Nikkei 225 4 Conclusion A Pareto-optimality Measures References Hierarchical Node Representation Learning for Stock Prediction 1 Introduction 2 Related Work 3 Method 3.1 Pairwise Attention Network 3.2 Hierarchical Node Matching 3.3 Multi-Granularity Representation Ensemble 4 Experiments 4.1 Experimental Settings 4.2 Performance Comparison 4.3 Ablation Study 4.4 Analysis 5 Conclusion References Application of APSO-BP Neural Network Algorithm in Stock Price Prediction 1 Introduction 2 BP Neural Network Algorithm 3 Improvement Strategy of Hybrid Algorithm 3.1 Coding Strategy and Calculation of Fitness Value 3.2 Particle Learning Strategy Based on BP Neural Network 4 A BP Neural Network Algorithm Integrating Adaptive PSO Algorithm (APSO-BP) 5 Empirical and Simulation 5.1 Data Selection 5.2 Topological Structure of BP Neural Network 5.3 Parameter Setting 5.4 Analysis of Empirical Results 6 Conclusions References The Research in Credit Risk of Micro and Small Companies with Linear Regression Model 1 Introduction 2 Linear Regression Model 2.1 Linear Normalization 2.2 Least Squares 2.3 Decision Tree 2.4 Confusion Matrix 3 Model Empirical Analysis 3.1 Data Processing 3.2 Enterprise Credit Scoring Model 3.3 Bank-to-Business Credit Strategy Model 4 Model Evaluation 4.1 Model Validation 4.2 Confusion Matrix 5 Conclusion References ICSI-Optimization Competition Deep-Layered Differential Evolution 1 Introduction 2 Background 2.1 DE 2.2 LSHADE 3 Deep-Layered Differential Evolution (DDE) 4 Experiments and Results 5 Conclusions References Dual-Populatuion Differential Evolution L-NTADE for ICSI-OC'2023 Competition 1 Introduction 2 Related Work 2.1 Differential Evolution 3 Proposed Approach 4 Experimental Setup and Results 5 Conclusion References Group Simulated Annealing Algorithm for ICSI-OC 2022 1 Introduction 2 Related Work 3 Detailed Description of the Algorithm 3.1 Group Strategy 3.2 Group Simulated Annealing Algorithm 4 Empirical Studies 4.1 Parameter Setting 4.2 Experimental Environment 4.3 Experimental Results 4.4 Convergence Contrastive Analysis 5 Conclusion and Future Work References Author Index