Bio-inspired computing . Part II : theories and applications : 16th international conference, BIC-TA 2021, Taiyuan, China, December 17-19, 2021 : revised selected papers / / Linqiang Pan [and three others] (editors)

• Pubblicazione: Singapore : , : Springer, , [2022]; ©2022
• Descrizione fisica: 1 online resource (448 pages)
• Disciplina: 006.38
• Soggetto topico: Natural computation; Artificial intelligence
• Persona (resp. second.): PanLinqiang
• Note generali: Includes index.
• Nota di contenuto: Intro -- Preface -- Organization -- Contents - Part II -- Contents - Part I -- Machine Learning and Computer Vision -- Point Clouds Registration Algorithm Based on Spatial Structure Similarity of Visual Keypoints -- 1 Introduction -- 2 Method -- 2.1 2D Visual Keypoints -- 2.2 Depth Completion and Back-Project -- 2.3 Screening 3D Keypoints -- 2.4 Rigid Transformation Parameter Estimation -- 3 Experiment -- 3.1 Data and Metrics -- 3.2 Experiment and Comparison -- 4 Conclusion -- References -- Software Defect Prediction Based on SMOTE-Tomek and XGBoost -- 1 Introduction -- 2 Related Work -- 2.1 Sampling Technique -- 2.2 Cost-Sensitive Learning -- 2.3 Ensample Learning Algorithm -- 3 The Proposed Model -- 3.1 SMOTE-Tomek -- 3.2 XGBoost -- 3.3 STX Model -- 4 Experiments -- 4.1 Datasets -- 4.2 Performance Measures -- 4.3 Results and Discussion -- 4.4 Statistical Comparison of Software Defect Predictors -- 5 Conclusion -- References -- Imbalance Classification Based on Deep Learning and Fuzzy Support Vector Machine -- 1 Introduction -- 2 Related Work -- 2.1 Data-Level -- 2.2 Algorithm-Level -- 3 Proposed Method -- 3.1 Feature Extraction with Deep Learning -- 3.2 Random Feature Oversampling Based on Center Distance -- 3.3 Fuzzy Support Vector Machine -- 4 Experiments and Results -- 4.1 Evaluation Metrics and Datasets -- 4.2 Experiment Settings -- 4.3 Results and Analysis -- 5 Conclusion -- References -- Community Detection Based on Surrogate Network -- 1 Introduction -- 2 Preliminaries -- 2.1 Spectral Clustering -- 2.2 EA-Based Community Detection -- 3 Proposed Method -- 4 Experiments -- 4.1 Experimental Setup -- 4.2 Experimental Results and Discussions -- 5 Conclusions -- References -- Fault-Tolerant Scheme of Cloud Task Allocation Based on Deep Reinforcement Learning -- 1 Introduction -- 2 Related Works -- 3 Cloud System and Fault Model. ; 3.1 Task Model -- 3.2 Fault Model -- 3.3 APSDQN MDP Model -- 4 APSDQN Implementation -- 5 Simulation Experiment -- 5.1 Experimental Setup -- 5.2 Experimental Results and Analysis -- 6 Conclusion -- References -- Attention-Guided Memory Model for Video Object Segmentation -- 1 Introduction -- 2 Related work -- 3 Methodology -- 3.1 Network Overview -- 3.2 Joint Attention Guider -- 3.3 Spatial-Temporal Feature Fusion -- 3.4 Implementation of Other Modules -- 3.5 Network Training -- 4 Experiments -- 4.1 Comparison to State-of-the-Art -- 4.2 Ablation Study -- 5 Conclusion -- References -- Multi-workflow Scheduling Based on Implicit Information Transmission in Cloud Computing Environment -- 1 Introduction -- 2 Workflow Schedule Model -- 2.1 Workflow Model -- 2.2 Problem Expression -- 3 Multifactorial Evolutionary Algorithm for DAG Schedule -- 3.1 MFEA Based on Combinatorial Population (CP-MFEA) -- 3.2 Generation of Population -- 3.3 Generation of Offspring -- 3.4 Evaluate Offspring -- 4 Experiment and Discuss -- 4.1 Basic Workflow Structure -- 4.2 Experimental Setup -- 4.3 Results and Analysis -- 5 Conclusion -- References -- Pose Estimation Based on Snake Model and Inverse Perspective Transform for Elliptical Ring Monocular Vision -- 1 Introduction -- 2 Ellipse Ring Contour Extraction Based on Snake Model -- 2.1 Rough Contour Extraction -- 2.2 Refined Contour Extraction -- 3 Ellipse Correction Based on Inverse Perspective Transformation -- 3.1 Solving Inverse Perspective Transformation Matrix -- 3.2 Ellipse Correction and Pose Estimation -- 4 Experimental Results and Analysis -- 5 Conclusion -- References -- Enhancing Aspect-Based Sentiment Classification with Local Semantic Information -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 4 Methodology -- 4.1 Embedding and Bidirectional LSTM -- 4.2 Obtaining Semantic Information. ; 4.3 FMDG: Obtaining Local Semantic Information -- 4.4 Information Fusion -- 4.5 Sentiment Classification -- 4.6 Training -- 5 Experiments -- 5.1 Dataset and Experiment Setup -- 5.2 Models for Comparison -- 5.3 Overall Result -- 5.4 Ablation Study -- 6 Conclusion -- References -- A Chinese Dataset Building Method Based on Data Hierarchy and Balance Analysis in Knowledge Graph Completion -- 1 Introduction -- 2 Problem Analysis -- 2.1 Existence of Meaningless Triples -- 2.2 Unbalanced Data Volume -- 3 Methods -- 3.1 Use Indicators to Measure Dataset Structure -- 3.2 Method of Constructing Chinese Dataset -- 3.3 Knowledge Graph Completion Model Selection -- 4 Experiments -- 5 Conclusions -- References -- A Method for Formation Control of Autonomous Underwater Vehicle Formation Navigation Based on Consistency -- 1 Introduction -- 2 AUV Formation System Description -- 2.1 Graph Theory -- 2.2 Kinematic Model -- 2.3 Information Interaction Model -- 3 Formation Control Algorithm Design -- 3.1 Definition of Covariate -- 3.2 Second-Order Consistency Control Algorithm -- 3.3 Model Predictive Control Rate Design -- 4 Simulation Research -- 4.1 Simulation Setup -- 4.2 Simulation Results and Analysis -- 5 Conclusion -- References -- A Formation Control Method of AUV Group Combining Consensus Theory and Leader-Follower Method Under Communication Delay -- 1 Introduction -- 2 Preliminaries and Modelling -- 2.1 Graph Theory -- 2.2 AUV Model -- 2.3 Communication Modelling -- 3 Consistency Control Algorithm Based on Leader-Following Method for AUV Group -- 3.1 Consistency Control Algorithm Without Communication Delay -- 3.2 Consistency Control Algorithm with Communication Delay -- 4 Simulation Results -- 4.1 Simulation of Consistency Control Algorithm Without Communication Delay -- 4.2 Simulation of Consistency Control Algorithm with Communication Delay -- 5 Conclusion. ; References -- stigLD: Stigmergic Coordination of Linked Data Agents -- 1 Introduction -- 2 Varieties of Stigmergy and Related Work -- 3 Process Algebra, Agents and Linked Systems -- 3.1 Theoretical Setting: CCS with Value-Passing -- 3.2 Linked Data Servers, Agents and Linked Systems -- 3.3 Synthesis -- 4 Stigmergy in Linked Systems -- 4.1 stigLD: A Domain Model for Stigmergic Linked Systems -- 4.2 stigFN: SPARQL Functions for Stigmergic Linked Systems -- 5 Use Case: Make-to-Order Fulfilment -- 5.1 Shop Floor Representation in StigLD -- 5.2 Agent Models -- 5.3 Evaluation -- 6 Conclusions and Future Work -- References -- An Approach for Optimal Coverage of Heterogeneous AUV Swarm Based on Consistency Theory -- 1 Introduction -- 2 Problem Formulation and Preliminaries -- 2.1 Voronoi Partition -- 2.2 Bioinspired Model -- 2.3 Region Partition Model -- 3 Optimal Region Partition with Consensus Protocol -- 3.1 Control Strategy Design -- 3.2 Consistency Analysis -- 4 Collaborative Target Search Based on Consensus Protocol -- 4.1 Collaborative Target Search Problem Modeling -- 4.2 Consistency Condition -- 5 Simulation Experiment and Analysis -- 5.1 Simulation of Optimal Region Partition -- 5.2 Simulation of Cooperative Target Search -- 6 Conclusion -- References -- An Improved Event-Triggered Control Method Based on Consistency Algorithm in Heterogeneous AUV Swarm Under Communication Delay -- 1 Introduction -- 2 Problem Description -- 2.1 Multi-AUV Swarm -- 2.2 Heterogeneous AUV Swarm -- 3 Controller Design -- 3.1 Heterogeneous AUV Swarm Consistency Control Algorithm Based on Event-Triggered -- 3.2 Heterogeneous AUV Group Consistency Algorithm Based on Event-Triggered Control Under Communication Time Delay -- 4 Simulation and Analysis -- 4.1 Simulation of Heterogeneous AUV Cluster Consistency Algorithm under Event Trigger Control. ; 4.2 Simulation of Heterogeneous AUV Cluster Consistency Algorithm with Time Delay Under Event Trigger Control -- 5 Conclusion -- References -- Edge Computing Energy-Efficient Resource Scheduling Based on Deep Reinforcement Learning and Imitation Learning -- 1 Introduction -- 2 Related Works -- 3 Scheduling System -- 3.1 Workload Processor -- 3.2 Problem Definition -- 3.3 Environment Model -- 4 Simulation Experiment -- 5 Conclusion -- References -- Metric Learning with Distillation for Overcoming Catastrophic Forgetting -- 1 Introduction -- 2 Related Work -- 2.1 Incremental Learning -- 2.2 Metric Learning -- 2.3 Knowledge Distillation -- 3 Proposed Method -- 3.1 Network Structure and its Losses -- 3.2 Knowledge Distillation Embedded in the Network -- 3.3 Classifier -- 4 Experiment -- 4.1 Datasets -- 4.2 Implementation Details -- 4.3 Ablation Experiment -- 4.4 Comparison Methods -- 5 Conclusions -- References -- Feature Enhanced and Context Inference Network for Pancreas Segmentation -- 1 Introduction -- 2 Related Work -- 2.1 Deep Learning Methods -- 2.2 Attention Mechanism -- 3 Methods -- 3.1 Feature Encoder -- 3.2 Feature Space Mapping -- 3.3 Feature Enhancement -- 3.4 Decoder -- 4 Experiment -- 4.1 Experiment Setup -- 4.2 Comparative Experiments -- 4.3 Comparative Experiments -- 5 Discussion and Conclusion -- References -- Object Relations Focused Siamese Network for Remote Sensing Image Change Detection -- 1 Introduction -- 2 Related Work -- 3 Methods -- 3.1 Basic Network Architecture -- 3.2 Geo-Objects Relations Module -- 3.3 Feature Enhancement -- 4 Experiment -- 4.1 Experiment Settings -- 4.2 Ablation Experiments -- 4.3 Comparative Experiments -- 5 Conclusion -- References -- MLFF: Multiple Low-Level Features Fusion Model for Retinal Vessel Segmentation -- 1 Introduction -- 2 Related Works -- 3 Multiple Low-Level Feature Fusion Model. ; 3.1 Multiple Low-Level Feature Extraction.
• Titolo autorizzato: Bio-Inspired Computing
• ISBN: 981-19-1253-X
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910556893003321
• Lo trovi qui: Univ. Federico II
• Serie: Communications in computer and information science ; ; Volume 1566.