top

  Info

  • Utilizzare la checkbox di selezione a fianco di ciascun documento per attivare le funzionalità di stampa, invio email, download nei formati disponibili del (i) record.

  Info

  • Utilizzare questo link per rimuovere la selezione effettuata.

Biblioteca

MARC Lista (tabellare)
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part IV
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part IV
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (508 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-59-1
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part IV -- CSDSE: Apply Cooperative Search to Solve the Exploration-Exploitation Dilemma of Design Space Exploration -- 1 Introduction -- 2 Background and Motivation -- 2.1 Design Space of Deep Neural Network Accelerator -- 2.2 Design Space Exploration Methods -- 2.3 Challenge of Exploration-Exploitation Dilemma -- 3 CSDSE Method -- 3.1 Profile of CSDSE -- 3.2 Definition of RL Framework -- 3.3 Cooperative Search and Heterogeneous Agents -- 3.4 Weighted Compact Buffer -- 4 Experiments and Analyses -- 4.1 Experiment Setup -- 4.2 Experiment Result and Analysis -- 5 Related Works -- 6 Conclusion -- References -- Differential Privacy in Federated Dynamic Gradient Clipping Based on Gradient Norm -- 1 Introduction -- 2 Related Work -- 3 Preliminary -- 3.1 Threat Model -- 3.2 Gradient Cropping -- 4 Differential Privacy in Federated Dynamic Gradient Clipping Based on Gradient Norm -- 4.1 Overall Framework -- 4.2 Relationship Between Gradient Variation and Clipped Threshold -- 4.3 Gradient Attenuation Clipped Threshold Based on Gradient Norm -- 4.4 Convergence Analysis -- 4.5 Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Setup -- 5.2 Performance of Privacy Protection -- 5.3 Global Model Availability -- 5.4 Universality of Clipped Thresholds -- 6 Conclusions -- References -- SolGPT: A GPT-Based Static Vulnerability Detection Model for Enhancing Smart Contract Security -- 1 Introduction -- 2 Problem Formulation -- 3 Related Work -- 3.1 Deep Learning-Based Approaches for Smart Contract Vulnerability Detection -- 3.2 GPT Models -- 4 SolGPT: Solidity Adapative GPT -- 4.1 SolTokenizer -- 4.2 Solidity Adaptive Pre-training -- 4.3 Vulnerability Detection Fine-Tunning -- 5 Experiment and Analysis -- 5.1 Comparisons with Traditional Approaches -- 5.2 Comparisons with Deep Learning Approaches.
5.3 Ablation Studies -- 6 Conclusion -- References -- Neuron Pruning-Based Federated Learning for Communication-Efficient Distributed Training -- 1 Introduction -- 2 Related Work -- 2.1 FL with Edge Computing -- 2.2 Communication Overhead Reduction in FL -- 3 Algorithm Description -- 3.1 Basic Model Selection -- 3.2 Model Pruning -- 4 Simulation Results and Analysis -- 4.1 Experimental Setting -- 4.2 Performance Analysis -- 4.3 Network Traffic Classification -- 5 Conclusion and Future Work -- References -- A Novel Network Topology Sensing Method for Network Security Situation Awareness -- 1 Introduction -- 2 Preliminaries and Related Works -- 2.1 Network Security Situation Awareness (NSSA) -- 2.2 Group Decision Making (GDM) -- 3 Network Topology Sensing Method for NSSA -- 3.1 Overview -- 3.2 Agent Model -- 3.3 Group Decision Making Based on Trusted Agents -- 4 Experimental Studies and Analysis -- 4.1 Prerequisites and Problem Descriptions -- 4.2 Experimental Process and Numerical Example -- 4.3 Comparisons and Discussions -- 5 Conclusions -- References -- Distributed Task Offloading Method Based on Federated Reinforcement Learning in Vehicular Networks with Incomplete Information -- 1 Introduction -- 2 Related Work -- 3 System Model -- 3.1 Local Computing -- 3.2 Edge Computing -- 4 Problem Formulation -- 5 Algorithm Design -- 5.1 Deep Q-Network -- 5.2 State Space -- 5.3 Action Space -- 5.4 Reward Function -- 5.5 Communication Mask -- 5.6 FDQN Algorithm -- 6 Performance Evaluation -- 6.1 Simulation Setup -- 6.2 Evaluation Results -- 7 Conclusion -- References -- Carbon Trading Based on Consortium Chain: Building, Modeling, and Analysis -- 1 Introduction -- 2 DC-Chain for Carbon Trading -- 2.1 Role Assignment -- 2.2 DC-Chain Construction and Smart Contract Design -- 3 Modeling Blockchain Performance -- 4 Analytical and Experimental Results.
4.1 Transaction Confirmation Time -- 4.2 Block Size -- 5 Conclusion -- References -- CAST: An Intricate-Scene Aware Adaptive Bitrate Approach for Video Streaming via Parallel Training -- 1 Introduction -- 2 Background and Motivation -- 2.1 ABR Algorithms -- 2.2 Scene Complexity -- 3 Design -- 3.1 CAST Overview -- 3.2 Neural Network Structure -- 3.3 Implementation -- 4 Evaluation -- 4.1 Methodology -- 4.2 Comparison with Existing ABR Algorithms -- 4.3 CAST vs. Existing ABR Approaches -- 4.4 Performance Under Different Network Environments -- 4.5 Performance Under Different Player Settings -- 4.6 Subjective Tests from Real Users -- 5 Conclusions -- References -- A Hybrid Active and Passive Cache Method Based on Deep Learning in Edge Computing -- 1 Introduction -- 2 Related Work -- 3 System Model -- 4 Proposed Methodology -- 4.1 Feature Extraction from Four Modalities -- 4.2 Multimodal Prediction Model -- 5 Experiment Setup -- 5.1 Dataset Setup -- 5.2 Evaluating Indicator -- 5.3 Baselines -- 6 Cached Updates -- 7 Evaluation Results -- 7.1 Simulation Setup -- 7.2 Evaluation Results -- 8 Conclusion -- References -- We Will Find You: An Edge-Based Multi-UAV Multi-Recipient Identification Method in Smart Delivery Services -- 1 Introduction -- 2 Related Work -- 3 Framework and Algorithms -- 3.1 Framework -- 3.2 Algorithms -- 4 Evaluation -- 4.1 Case Study -- 4.2 Experiments -- 5 Conclusions and Future Work -- References -- Multi-objective Optimization for Joint Handover Decision and Computation Offloading in Integrated Communications and Computing 6G Networks -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulations -- 3.1 System Model -- 3.2 Communication Model -- 3.3 Computation Model -- 3.4 Problem Formulation -- 4 Joint Handover Decision and Computation Offloading Method Based on MOO -- 4.1 The Flowchart of Mobility Management in CCN.
4.2 Trigger Mechanisms of Handover and Offloading -- 4.3 Joint Decision Method Based on MOO -- 4.4 The Convergence Analysis of JD-MOO Method -- 5 Performance Evaluation -- 5.1 Experiments Setting -- 5.2 Performance of Handover Decision -- 5.3 Performance of Computation Offloading -- 5.4 Analysis of Services Satisfaction Degree -- 5.5 Analysis of Running Time and Convergence -- 6 Conclusion -- References -- A Joint Optimization Scheme in Heterogeneous UAV-Assisted MEC -- 1 Introduction -- 2 Related Work -- 2.1 MEC -- 2.2 UAV Scheduling -- 2.3 Task Allocation -- 3 System Model -- 3.1 Problem Statement -- 3.2 System Overview -- 3.3 UAV Scheduling Model -- 3.4 Task Execution Model -- 3.5 UAV Hovering Model -- 3.6 Problem Formulation -- 4 Proposed Approach -- 4.1 UAV Scheduling Algorithm -- 4.2 Task Allocation Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Settings -- 5.2 Experimental Results -- 6 Conclusion -- References -- Multi-UAV Collaborative Face Recognition for Goods Receiver in Edge-Based Smart Delivery Services -- 1 Introduction -- 2 Framework Design and Workflow -- 2.1 Framework Overview -- 2.2 Framework Execution Process -- 3 Related Technologies and Algorithms -- 3.1 Face Detection -- 3.2 Face Fusion -- 3.3 Face Recognition -- 4 Experiments and Evaluation -- 4.1 Case Studies -- 4.2 Evaluation -- 5 Conclusions -- References -- Distributed Generative Adversarial Networks for Fuzzy Portfolio Optimization -- 1 Introduction -- 2 Generative Adversarial Networks Architecture -- 2.1 Theoretical Background -- 2.2 Architecture -- 2.3 Generator -- 2.4 Discriminator -- 2.5 Adversarial Training -- 3 Generative Adversarial Networks Architecture -- 4 Experimental Results -- 4.1 Experimental Setting -- 4.2 Experimental Results -- 4.3 Experimental Evaluation -- 5 Parallel Fuzzy Portfolio Optimization -- 5.1 Fuzzy Mean-CVaR Portfolio Model.
5.2 Optimization Algorithms -- 5.3 Parallelization and Results -- 6 Conclusion -- References -- A Novel Transaction Processing Model for Sharded Blockchain -- 1 Introduction -- 2 Related Work -- 2.1 Blockchain Sharding Protocols -- 2.2 Transaction Processing -- 3 Proposed Model -- 3.1 Overview of Model -- 3.2 Threat Model -- 4 Transaction Admission Control Algorithm -- 4.1 Problem Formulation -- 4.2 Algorithm Design -- 5 Transaction Verification Design -- 5.1 Verification Based on TSS -- 5.2 Verification Peers Selection -- 5.3 Transaction Dependency Handling -- 6 Experiment and Evaluation -- 6.1 Implementation -- 6.2 Performance Evaluation -- 7 Conclusion -- References -- Computation Offloading Based on Deep Reinforcement Learning for UAV-MEC Network -- 1 Introduction -- 2 System Model and Problem Description -- 2.1 System Model -- 2.2 Communication Model -- 2.3 Task Calculation Model -- 2.4 Problem Formulation -- 3 Proposed Scheme -- 3.1 SDQN Algorithm -- 3.2 Reinforcement Learning Modeling -- 3.3 Training and Testing -- 4 Simulation Results -- 4.1 Simulation Setting -- 4.2 Simulation Results and Discussions -- 5 Conclusion -- References -- Privacy-Aware Scheduling Heuristic Based on Priority in Edge Environment -- 1 Introduction -- 2 Related Work -- 3 Problem Modeling -- 3.1 Application Model -- 3.2 Resource and Processing Time Model -- 4 Method -- 4.1 Task Selection Phase -- 4.2 Processor Selection Phase -- 5 Evaluation -- 5.1 Settings -- 5.2 The Performance Comparison Experiment for Privacy Application -- 5.3 The Performance Comparison Experiment for General Application -- 6 Conclusion -- References -- A Novel Multi-objective Evolutionary Algorithm Hybrid Simulated Annealing Concept for Recommendation Systems -- 1 Introduction -- 2 Related Work -- 3 Problem Definition -- 3.1 Multi-objective Evolutionary Algorithm -- 4 The Proposed Model.
4.1 Objective Functions.
Record Nr. UNINA-9910841857303321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part V
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part V
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (525 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-08-7
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part V -- EQFF: An Efficient Query Method Using Feature Fingerprints -- 1 Introduction -- 2 Related Work -- 2.1 Privacy of Bloom Filters -- 2.2 Fuzzy Query -- 3 System Overview -- 4 Construction of Data Feature Fingerprints -- 4.1 Construction and Chunking of Data Feature Fingerprints -- 4.2 Fingerprint Compression Storage Algorithm -- 5 Precise Query Library and Rapid Query -- 5.1 Acceleration Algorithm Based on Inverted Index Library -- 5.2 Precise Query Algorithm Based on Bloom Filter -- 6 Fuzzy Query and Similarity Measurement -- 6.1 Similarity Measurement Based on Counting Bloom Filters -- 6.2 Search Algorithm Acceleration Using Parallel Computing -- 7 Evaluation -- 7.1 Experimental Settings and Dataset -- 7.2 Storage Overhead and Parameter Selection for Feature Fingerprints -- 7.3 Performance for Precise Query -- 7.4 Performance for Fuzzy Query -- 8 Conclusion -- References -- Joint Controller Placement and Flow Assignment in Software-Defined Edge Networks -- 1 Introduction -- 2 Related Work -- 2.1 Assignment Schemes -- 2.2 Placement and Assignment Schemes -- 2.3 Placement and Assignment Schemes in SDN-Enabled Edge Network -- 3 System Model and Problem Formulation -- 3.1 System Model -- 3.2 Flow-Setup Delay Model -- 3.3 Problem Formulation -- 4 Controller Placement and Flow Assignment -- 4.1 Controller Placement -- 4.2 Flow Assignment -- 5 Experiment and Performance Evaluation -- 5.1 Controller Response Time -- 5.2 High Queuing Delay Flows -- 5.3 Folw-Setup Delay Constraint Violated Flows -- 5.4 Algorithm Execution Time -- 6 Conclusion -- References -- Distributed Latency-Efficient Beaconing for Multi-channel Asynchronous Duty-Cycled IoT Networks -- 1 Introduction -- 2 Related Works -- 3 Problem Definition -- 4 Distributed Latency-Efficient Scheduling Algorithm.
4.1 Special Structures for Collision Avoidance -- 4.2 Computing Candidate Node Transmission Schedule -- 4.3 The Distributed Beaconing Schedule Algorithm -- 5 Performance Analysis -- 6 Simulation Results -- 6.1 Performance Comparison in Single-channel Duty-Cycled Scenarios -- 6.2 Performance Analysis in Multi-channel Duty-Cycled Scenarios -- 7 Conclusions -- References -- Improved Task Allocation in Mobile Crowd Sensing Based on Mobility Prediction and Multi-objective Optimization -- 1 Introduction -- 2 Related Work -- 3 System Overview and Problem Formulation -- 3.1 System Overview -- 3.2 Problem Formulation -- 3.3 Hardness Analysis of M3P Problem -- 4 Fuzzy Control System -- 4.1 Fuzzy Set and Membership Function -- 4.2 Fuzzy Inference Engine -- 4.3 COG Defuzzification -- 5 Task Allocation Algorithm -- 5.1 Most Possible First Algorithm -- 5.2 Improved NSGA-II Algorithm -- 6 Evaluation -- 6.1 Dataset -- 6.2 Baselines -- 6.3 Performance of Mobility Prediction -- 6.4 Performance of Task Allocation -- 7 Conclusion -- References -- SMCoEdge: Simultaneous Multi-server Offloading for Collaborative Mobile Edge Computing -- 1 Introduction -- 2 Related Work -- 2.1 Edge-Cloud Collaboration -- 2.2 Edge-Edge Collaboration -- 3 System Model and Problem Formulation -- 3.1 System Model -- 3.2 Problem Formulation -- 4 Proposed Approach -- 4.1 SMCoEdge Framework -- 4.2 Improved DQN for Multi-ES Selection -- 4.3 Closed-Form Solution for Task Allocation -- 4.4 DRL-SMO Algorithm -- 5 Performance Evaluation -- 5.1 Setting -- 5.2 Baselines -- 5.3 Results -- 6 Conclusions -- References -- Image Inpainting Forensics Algorithm Based on Dual-Domain Encoder-Decoder Network -- 1 Introduction -- 2 Related Works -- 2.1 Image Inpainting Forensics -- 2.2 Attention Mechanism -- 3 Proposed Method -- 3.1 S-Encoder -- 3.2 F-Encoder -- 3.3 CMAF -- 3.4 Attention-Gated Decoder Network.
3.5 Loss Function -- 4 Experiment -- 4.1 Experimental Setup -- 4.2 Comparing with Previous Methods -- 4.3 Abation Study -- 4.4 Robustness Evaluation -- 5 Conclusions -- References -- Robust Medical Data Sharing System Based on Blockchain and Threshold Rroxy Re-encryption -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Blockchain -- 3.2 Threshold Proxy Re-encryption -- 4 BMDSS Model -- 4.1 System Model -- 4.2 Application -- 5 System Evaluation -- 5.1 Environment -- 5.2 Data Storage Performance -- 5.3 Data Sharing Performance -- 5.4 Security Analysis -- 6 Conclusion -- References -- FRAIM: A Feature Importance-Aware Incentive Mechanism for Vertical Federated Learning -- 1 Introduction -- 2 Related Work -- 2.1 Federated Learning -- 2.2 Incentive Mechanism of Federated Learning -- 3 System Model and Problem Definition -- 3.1 Vertical Federated Learning -- 3.2 System Model -- 3.3 Problem Definition -- 4 Incentive Mechanism Design of Vertical Federated Learning -- 4.1 Data Synthesis and Feature Importance Estimation -- 4.2 Design of FRAIM -- 4.3 Theoretical Analysis -- 5 Experimental Evaluation -- 5.1 Experiment Setup -- 5.2 Performance on Three Datasets -- 5.3 Performance with Different Budgets -- 6 Conclusion -- References -- A Multi-objective Method for Energy-Efficient UAV Data Collection Communications -- 1 Introduction -- 2 Models and Preliminaries -- 2.1 System Model -- 2.2 Network Model -- 2.3 Energy Consumption Model of the UAV -- 3 Problem Formulation and Analysis -- 3.1 Problem Statement -- 3.2 Optimization Objectives -- 4 Algorithm -- 4.1 Conventional MOGOA -- 4.2 EMOGOA -- 5 Simulation Results and Analysis -- 5.1 Simulation Setups -- 5.2 Simulation Results -- 6 Conclusion -- References -- Black-Box Graph Backdoor Defense -- 1 Introduction -- 2 Related Work -- 2.1 Graph Neural Network Backdoor -- 2.2 Black Box Backdoor Defenses.
3 Methodology -- 3.1 Defense Assumptions and Goals -- 3.2 Key Intuition -- 3.3 BloGBaD -- 4 Experiment -- 4.1 Experiment Settings -- 4.2 Evaluation Metrics -- 4.3 GTA Defense -- 4.4 GBA Defense -- 5 Conclusion -- References -- DRL-Based Optimization Algorithm for Wireless Powered IoT Network -- 1 Introduction -- 2 Related Work -- 2.1 Binary Offloading -- 2.2 Partial Offloading -- 3 System Model and Problem Formulation -- 3.1 System Model -- 3.2 Local Computing Mode -- 3.3 Edge Computing Mode -- 3.4 Problem Formulation -- 4 Deep-Reinforcement-Learning-Based Offloading Algorithm -- 4.1 Deep-Reinforcement-Learning-Based Method for Solving Top-Problem -- 4.2 Mixed Optimization Method for Solving Sub-problem -- 5 Numerical Results -- 5.1 Convergence Performance -- 5.2 Influence of Different Parameters on Result -- 5.3 Numerical Evaluation -- 6 Conclusion -- References -- Efficiently Running SpMV on Multi-core DSPs for Banded Matrix -- 1 Introduction -- 2 Background -- 2.1 FT-M7032 Heterogeneous Processor -- 2.2 Sparse Matrix Storage Format -- 2.3 Banded Matrix -- 3 Our Approach -- 3.1 Architecture Analysis -- 3.2 SpMV_Band Design and Implementation -- 4 Performance Results -- 4.1 Experimental Setup -- 4.2 Performance Evaluation -- 5 Related Work -- 6 Conclusion -- References -- SR-KGELS: Social Recommendation Based on Knowledge Graph Embedding Method and Long-Short-Term Representation -- 1 Introduction -- 2 Related Work -- 3 Problem Definition -- 4 The Proposed Model -- 4.1 Embedding Layer and Input Data Pre-processing -- 4.2 User Modeling -- 4.3 Item Modeling -- 4.4 Relative Location Difference Modeling -- 4.5 Model Training -- 5 Experiments -- 5.1 Experimental Settings -- 5.2 Comparative Results -- 5.3 Ablation Study -- 5.4 Analysis of Parameters -- 6 Conclusion -- References.
CMMR: A Composite Multidimensional Models Robustness Evaluation Framework for Deep Learning -- 1 Introduction -- 2 Attacks, Defense, and Metrics -- 2.1 Attack Methods -- 2.2 Defense Methods -- 2.3 Comparison Metrics -- 3 Composite Multidimensional Model Robust Evaluation Method -- 3.1 Motivation -- 3.2 CMMR Framework -- 4 Experimental Analysis for CMMR -- 4.1 3-Dimension Models Robustness Evaluation. -- 4.2 Evaluating the Robustness of Two Sets of Models by CMMR -- 5 Conclusion -- References -- Efficient Black-Box Adversarial Attacks with Training Surrogate Models Towards Speaker Recognition Systems -- 1 Introduction -- 2 Related Work -- 2.1 White-Box and Gray-Box Attacks -- 2.2 Black-Box Attacks -- 3 Methodology -- 3.1 Problem Description -- 3.2 The Proposed Framework -- 3.3 Surrogate Models -- 3.4 The Training of Surrogate Model -- 3.5 Adversarial Examples Generation -- 4 Experiments -- 4.1 Experimental Settings -- 4.2 Transferability Analysis -- 4.3 Frequency Analysis -- 4.4 Performance Comparison -- 5 Discussion of Potential Defenses -- 6 Conclusions and Future Work -- References -- SW-LeNet: Implementation and Optimization of LeNet-1 Algorithm on Sunway Bluelight II Supercomputer -- 1 Introduction -- 2 Related Work -- 2.1 Convolutional Neural Network -- 2.2 The Works Based on the Sunway Supercomputers -- 3 The Sunway Bluelight II Supercomputer and LeNet-1 -- 3.1 The Architecture of the Sunway Bluelight II Supercomputer -- 3.2 The Architecture of LeNet-1 -- 3.3 Serial Execution Mode of LeNet-1 -- 4 SW-LeNet - Parallel Optimization of LeNet-1 on the Sunway Bluelight II Supercomputer -- 4.1 Thread-Level Optimization -- 4.2 Process-Level Optimization -- 5 Evaluation -- 5.1 The Performance of Thread-Level Optimization -- 5.2 The Performance of Process-Level Optimization -- 5.3 The Performance of SW-LeNet -- 5.4 Scaling -- 6 Conclusion -- References.
Multi-label Detection Method for Smart Contract Vulnerabilities Based on Expert Knowledge and Pre-training Technology.
Record Nr. UNINA-9910841859103321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VI
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VI
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (512 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-11-7
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part VI -- An Uncertainty-Aware Auction Mechanism for Federated Learning -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Definition -- 3.1 Federated Learning -- 3.2 Auction Framework -- 3.3 Problem Definition -- 4 The VCG-Based Auction Mechanism -- 4.1 Problem Transformation -- 4.2 Expected Social Welfare Maximization Algorithm -- 4.3 Payment -- 4.4 Properties -- 5 Greedy Algorithm Based Auction Mechanism -- 5.1 Choice of Clients' Training Data -- 5.2 Payment -- 5.3 Properties -- 5.4 Further Option for Reducing Time Complexity -- 6 Simulation Results -- 6.1 Experiment Settings -- 6.2 Experiment Analysis -- 7 Conclusion -- References -- Persistent Sketch: A Memory-Efficient and Robust Algorithm for Finding Top-k Persistent Flows -- 1 Introduction -- 2 Related Work -- 3 Problem Statement -- 4 The PE-Sketch Design -- 4.1 Approach Overview -- 4.2 Event Sampling Based on Flip Filter -- 4.3 Candidate Tracking -- 5 Performance Analysis -- 5.1 Parameters of Flip Filter -- 5.2 Estimation Error Bound of Candidate Matrix -- 6 Experimental Evaluation -- 6.1 Experiment Setup -- 6.2 Evaluation on Event Sampling -- 6.3 Finding Top-k Persistent Flows -- 6.4 Parameter Setting -- 7 Conclusion -- References -- FaCa: Fast Aware and Competition-Avoided Balancing for Data Center Network -- 1 Introduction -- 2 Related Work -- 2.1 Centralized Load Balancing -- 2.2 Distributed Load Balancing -- 3 Background -- 4 Motivation -- 4.1 Why Not ECMP? -- 4.2 A Motivation Example -- 5 Design -- 5.1 Overview -- 5.2 Network Model -- 5.3 Partly Global Vision with INT -- 5.4 Load Estimate -- 5.5 Flowing& -- Jumping -- 6 Implementation -- 6.1 Productive Experiment Setting -- 6.2 Implementation Challenges -- 7 Evaluation -- 7.1 Throughput Performance -- 7.2 Concurrent Competition -- 7.3 Micro-benchmark -- 8 Conclusion.
References -- Optimizing GNN Inference Processing on Very Long Vector Processor -- 1 Introduction -- 2 Background -- 2.1 GNN -- 2.2 Brief Introduction to the Long Vector Architecture Processor -- 3 Related Works -- 4 Efficient GNNs Inference on Long Vector Architecture -- 4.1 Efficient GNN Operator Library -- 4.2 GNN Operator Fusion Strategy -- 4.3 Heterogeneous Multi-threaded Cooperative Strategy for Long Vector Architecture -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Overall Performance -- 5.3 Operator Fusion Strategy -- 5.4 Heterogeneous Multi-threaded Cooperative Strategy -- 6 Conclusion -- References -- GDTM: Gaussian Differential Trust Mechanism for Optimal Recommender System -- 1 Introduction -- 2 Related Work -- 2.1 Trust-Based Recommendation -- 2.2 Recommended Protection Mechanism -- 3 Trade-Off Mechanism -- 3.1 Trust-Related Definition -- 3.2 Trust Recommendation -- 3.3 Gaussian Differential Trust Mechanism -- 4 Theoretical Analysis -- 5 Experiment Summary -- 5.1 Research Questions -- 5.2 Benchmark Model -- 5.3 Evaluation Metrics -- 5.4 Performance Analysis -- 6 Conclusion -- References -- Privacy-Enhanced Dynamic Symmetric Searchable Encryption with Efficient Searches Under Sparse Keywords -- 1 Introduction -- 2 Preliminaries -- 2.1 Notations -- 2.2 Dynamic Searchable Symmetric Encryption -- 3 Forward Privacy DSSE: SGsse-F -- 3.1 Construction of SGsse-F -- 3.2 An Example of SGsse-F -- 3.3 Correctness -- 3.4 Security Analysis -- 4 Forward and Backward Privacy DSSE: SGsse-FB -- 4.1 Correctness -- 4.2 Security Analysis -- 5 Related Works -- 6 Benefits and Drawbacks of DSSE Schemes Analysis -- 7 Evaluation -- 7.1 Experimental Setup -- 7.2 Experimental Analysis -- 8 Conclusion -- References -- CBA Sketch: A Sketching Algorithm Mining Persistent Batches in Data Streams -- 1 Introduction -- 1.1 Background and Motivation -- 1.2 Our Solution.
1.3 Key Contributions -- 2 Problem Formulation and Related Work -- 2.1 Problem Formulation -- 2.2 Related Work -- 3 The CBA Sketch -- 3.1 Circular-Time Sketch -- 3.2 Bloom Filter-Based Existence Recorder -- 3.3 Approximate Size Recorder -- 3.4 Query -- 4 Evaluation -- 4.1 Strawman Approach -- 4.2 Experimental Setup -- 4.3 Sensitivity Analysis -- 4.4 Experiments on ARE -- 4.5 Experiments on Throughput -- 5 Conclusion -- References -- Joint Optimization of System Bandwidth and Transmitting Power in Space-Air-Ground Integrated Mobile Edge Computing -- 1 Introduction -- 2 Related Works -- 2.1 Research on Air-Ground Communication -- 2.2 Research on Space-Ground Communication -- 2.3 Research on SAGIN Communication -- 3 System Model and Problem Formulation -- 3.1 Network Architecture -- 3.2 Satellite Coverage Time -- 3.3 IoRT Devices-Satellite/UAV Channel -- 3.4 Traffic Offloading Delay -- 3.5 Problem Formulation -- 4 Algorithm Design -- 4.1 DRL-Oriented Problem Transformation -- 4.2 SAC-Based Offloading Algorithm -- 5 Performance Evaluation -- 5.1 Parameter Setting -- 5.2 Performance Evaluation -- 6 Conclusion -- References -- A Novel Sensor Method for Dietary Detection -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Coupling Effect -- 3.2 Hardware Design -- 3.3 Event Detection Method Based on SVM -- 3.4 Chewing Counting Algorithm -- 4 Evaluation and Discussion -- 4.1 Experimental Methodology -- 4.2 The Evaluation of Chewing Event Detection -- 4.3 The Evaluation of Chewing Counting -- 5 Conclusion -- References -- Accelerated Optimization for Simulation of Brain Spiking Neural Network on GPGPUs -- 1 Introduction -- 2 Related Work -- 2.1 Spiking Neural Networks -- 2.2 SNN Simulation Tools -- 2.3 CUDA Programming on Multiple GPUs -- 3 Method -- 3.1 Workload-Oriented Partitioning -- 3.2 Memory Access Optimization.
3.3 Branch-Matched Thread Organization -- 3.4 Spike Synchronization Optimization -- 4 Experiments -- 4.1 Setup -- 4.2 Simulation Results -- 5 Conclusion -- References -- An Improved GPU Acceleration Framework for Smoothed Particle Hydrodynamics -- 1 Introduction -- 2 SPH Method -- 3 GPU Computation Using CUDA -- 3.1 Data Layout -- 3.2 CUDA Implementation -- 3.3 CUDA Acceleration -- 4 Results -- 4.1 GPU Simulation -- 4.2 Performance Evaluation -- 5 Conclusion and Future Work -- References -- MDCF: Multiple Dynamic Cuckoo Filters for LSM-Tree -- 1 Introduction -- 2 Background and Motivation -- 2.1 LSM-Tree Based KV Store -- 2.2 Serious Read Amplification of LSM-Tree -- 2.3 Motivation -- 3 Proposed Approach -- 3.1 The Design of MDCF -- 3.2 Maintenance of MDCF -- 3.3 The Flow of Read Operations -- 4 Evaluation -- 4.1 Experiment Setup -- 4.2 Workloads -- 4.3 Performance Evaluation -- 5 Related Work -- 6 Conclusion -- References -- A Game Theory Based Task Offloading Scheme for Maximizing Social Welfare in Edge Computing -- 1 Introduction -- 2 Motivation and Related Works -- 2.1 Motivation -- 2.2 Related Works -- 3 System Model -- 3.1 User Energy Consumption Model -- 3.2 User Utility Model -- 4 Stackelberg Game Based Offloading Algorithm -- 5 Evaluation -- 5.1 Evaluation Settings -- 5.2 Results -- 6 Conclusion -- References -- Research on Dos Attack Simulation and Detection in Low-Orbit Satellite Network -- 1 Introduction -- 1.1 A Subsection Sample -- 2 Related Work -- 3 Simulating DoS Attacks in Low Earth Orbit Satellite Networks -- 3.1 Satellite Physical Link Simulation Based on STK -- 3.2 Network Simulation Platform Based on OVS -- 3.3 SDN Control Platform Based on RYU -- 3.4 Network Attack Deployment Based on Docker Containers -- 4 Building and Detecting Datasets for Satellite Network Attacks -- 4.1 Overall Framework -- 4.2 Dataset Construction.
4.3 Stacking-Based Attack Detection -- 5 Experimental Results -- 6 Conclusion -- References -- Malware Detection Method Based on Visualization -- 1 Introduction -- 2 Related Work -- 2.1 Malware and Detection Methods -- 2.2 Visualization -- 2.3 Ensemble Machine Learning -- 3 Visualization-Based Malware Detection -- 3.1 Method Framework -- 3.2 Specific Implementation -- 4 Experiments and Evaluations -- 4.1 Data Set and Experimental Environment -- 4.2 Experimentation -- 5 Conclusion -- References -- TOC: Joint Task Offloading and Computation Reuse in Vehicular Edge Computing -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulation -- 3.1 System Model -- 3.2 Problem Formulation -- 4 TOC Framework -- 4.1 Computational Reuse -- 4.2 Minimum Weighted Vertex Search Algorithm -- 5 Experiment and Evaluation -- 5.1 Effects of Different Numbers of Tasks on Completion Time and Computational Execution Time on RSUs -- 5.2 Effects of Different Computation Capabilities on Completion Time and Computational Execution Time on RSUs -- 5.3 Lookup Rate Performance -- 6 Conclusion -- References -- Distributed Task Offloading for IoAV Using DDP-DQN -- 1 Introduction -- 2 Related Work -- 3 System Model -- 3.1 System Overview -- 3.2 Transmission Model -- 3.3 Task Delay -- 3.4 Energy Consumption Model -- 3.5 Formulation of Task Offloading Problem -- 4 DDP-DQN Algorithm in IoAV Offloading -- 4.1 IoAV Environment Model -- 4.2 DDP-DQN Based Task Offloading Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Settings -- 5.2 Results and Analysis -- 6 Conclusion -- References -- A Task Offloading and Resource Allocation Optimization Method in End-Edge-Cloud Orchestrated Computing -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulation -- 3.1 Network Model -- 3.2 Communication Model -- 3.3 Computation Model -- 3.4 Problem Formulation.
4 DRL-Based Task Offloading and Resource Allocation in EECOC.
Record Nr. UNINA-9910841862403321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part III
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part III
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (524 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9707-98-6
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part III -- Addressing Coupled Constrained Reinforcement Learning via Interative Iteration Design -- 1 Introduction -- 2 Preliminaries -- 2.1 Coupled Constrained Markov Decision Process -- 2.2 Policy Gradient -- 3 InIt for Decoupling High-Dimensional Coupled Action Space -- 3.1 InIt Design -- 3.2 InIt-Based Solution -- 3.3 Extension to High-Dimensional Action Space -- 4 Experiments -- 4.1 Experiment Setups -- 4.2 Baselines -- 4.3 Performance Evaluation -- 5 Conclusion and Future Work -- References -- A Path Planning and Obstacle Avoidance Method for USV Based on Dynamic-Target APF Algorithm in Edge -- 1 Introduction -- 2 Related Works -- 3 System Design and Environment Modeling -- 3.1 System Overview -- 3.2 Environment Modeling -- 4 Algorithm Design -- 4.1 Improved A* Algorithm -- 4.2 Improved APF Algorithm -- 4.3 DTAPF Method -- 4.4 Offset Guidance -- 5 Evaluation -- 5.1 Simulation Environment Configurations -- 5.2 Performance of Algorithms -- 5.3 Verification of System Architecture -- 6 Conclusion -- References -- Node-Disjoint Paths in Balanced Hypercubes with Application to Fault-Tolerant Routing -- 1 Introduction -- 2 Preliminaries -- 3 Node-Disjoint Paths -- 3.1 The Mixed Sequence -- 3.2 Construction of Node-Disjoint Paths -- 4 Simulations and Performance Evaluation -- 5 Conclusion -- References -- Performance Evaluation of Spark, Ray and MPI: A Case Study on Long Read Alignment Algorithm -- 1 Introduction -- 2 Methods -- 2.1 Parallel Algorithm Design -- 2.2 Index Sharing -- 2.3 Sequence Segmentation -- 2.4 Parallel Alignment -- 2.5 Asynchronous Output -- 3 Results -- 3.1 Experimental Setup -- 3.2 Input Data -- 3.3 Results Consistency -- 3.4 Strong Scalability -- 3.5 Weak Scalability and Peak RAM Memory -- 3.6 Efficiency -- 3.7 Throughput -- 3.8 Latency -- 3.9 Load Balance -- 4 Related Work.
5 Discussion -- 6 Conclusion -- References -- Fairness Analysis and Optimization of BBR Congestion Control Algorithm -- 1 Introduction -- 2 Related Work -- 3 Problem Analysis -- 3.1 RTT Fairness in BBR Algorithm -- 4 Algorithm Optimization -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Results on RTT Fairness -- 6 Conclusion -- References -- Accelerating QUIC with AF_XDP -- 1 Introduction -- 2 Background -- 2.1 QUIC Acceleration -- 2.2 Network Stack Optimization -- 2.3 eBPF and AF_XDP -- 2.4 AF_XDP Datapath -- 3 Design and Implementation -- 3.1 Overview -- 3.2 Architecture -- 3.3 Implementation -- 4 Evaluation -- 4.1 Experiment Setup -- 4.2 QUIC Performance -- 4.3 CPU Usage -- 4.4 Network Policy -- 5 Discussion -- 6 Conclusion -- References -- Segmenta: Pipelined BFT Consensus with Slicing Broadcast -- 1 Introduction -- 2 Related Work -- 3 Analysis and Design -- 3.1 Directly Apply RBC Module in Hotstuff -- 3.2 Design to Reduce the Additional Communication Round -- 4 System Model -- 5 The Segmenta Protocol -- 5.1 Segmenta-Specific Data Structures -- 5.2 Segmenta Phases -- 5.3 Safety, Liveness, and Communication Complexity -- 5.4 Echoing Shards Optimistically -- 6 Chained Segmenta Protocol -- 7 Evaluation -- 7.1 Experimental Setup -- 7.2 Prediction Model -- 7.3 Throughput and Latency -- 8 Conclusion -- References -- Synthetic Data Generation for Differential Privacy Using Maximum Weight Matching -- 1 Introduction -- 2 Related Work -- 3 Preliminary -- 3.1 Problem Statement -- 3.2 Differential Privacy -- 4 Our Solution -- 4.1 Construction of Attribute Graph -- 4.2 Marginal Selection -- 4.3 PostProcess -- 4.4 Data Synthesis -- 5 Experiments -- 5.1 Experimental Settings -- 5.2 Experimental Results -- 6 Conclusion -- A Proof of Lemma -- References -- Approximate Multicast Coflow Scheduling in Reconfigurable Data Center Networks -- 1 Introduction.
2 Problem Formulation -- 2.1 System Model -- 2.2 Circuit Scheduling Model -- 2.3 Problem Formulation -- 3 The Delay-Efficient Multicast Coflow Scheduling Algorithm -- 3.1 Multicast Coflow Scheduling -- 3.2 Circuit Scheduling -- 3.3 Approximation Ratio Analysis -- 4 Numerical Evaluation -- 4.1 Simulation Settings -- 4.2 Performance Evaluation -- 5 Related Work -- 6 Conclusion -- References -- DAS: A DRAM-Based Annealing System for Solving Large-Scale Combinatorial Optimization Problems -- 1 Introduction -- 2 Background -- 2.1 Ising Model -- 2.2 The Overview of DRAM -- 2.3 In-DRAM Bulk Bitwise Operations -- 3 DAS: Detailed Design -- 3.1 An Overview of DAS -- 3.2 Spin Propagation Between Spin Nodes -- 3.3 Calculation of Local Search Term -- 3.4 Design of Probability Flipping -- 4 Evaluation and Applications -- 4.1 Area Overhead -- 4.2 Application and Analysis -- 5 Conclusion -- References -- Graph Neural Network for Critical Class Identification in Software System -- 1 Summary -- 2 Related Works -- 2.1 Research on the Importance of Nodes in Complex Networks -- 2.2 Research on Key Node Identification in Software Network -- 3 Basic Theory -- 3.1 Node Feature Extraction Based on Direct Extrapolation Graph Neural Network -- 3.2 Node Feature Extraction Based on Inductive Graph Neural Network -- 4 Research methods -- 4.1 Modeling Class dependency network -- 4.2 Network Embedding Learning -- 4.3 Key Class Identification -- 5 Experimental Analysis -- 5.1 Experiment Data -- 5.2 Experiment Setup -- 5.3 Analysis of Experimental Results -- 6 Conclusion -- References -- Spatio-Temporal Fusion Based Low-Loss Video Compression Algorithm for UAVs with Limited Processing Capability -- 1 Introduction -- 2 System Overview -- 3 Algorithm Design -- 3.1 Feature Clustering Based Temporal Video Sampling -- 3.2 Dynamic Encoding Based Spatial Video Sampling -- 4 Evaluation.
4.1 Experimental Setup -- 4.2 Experimental Results -- 5 Implementation -- 6 Related Work -- 7 Conclusion -- References -- CRAFT: Common Router Architecture for Throughput Optimization -- 1 Introduction -- 2 Related Work -- 3 The Proposed Viaduct Theory in Throughput Optimization -- 3.1 Viaduct Theory in Throughput Optimization -- 3.2 Comparison of Ring Decouple Theory and Viaduct Theory in Low-Cost Router -- 4 A Common Router Architecture for Throughput Optimization Based on Viaduct Theory -- 4.1 Flow Control Optimization with Viaduct Theory -- 4.2 Pipeline Stage and Latency Optimization -- 4.3 High Throughput Optimization with Viaduct Theory -- 4.4 Low Hardware Cost with Viaduct Theory -- 4.5 Bypass Strategy for Low Workload -- 5 Experiment Results -- 5.1 For Low-Cost Program -- 5.2 For High-Performance Program -- 5.3 Hardware Overhead -- 6 Conclusion -- References -- A Cross-Chain System Supports Verifiable Complete Data Provenance Queries -- 1 Introduction -- 2 Related Work -- 2.1 Query Function -- 2.2 Verifiable Query -- 2.3 Cross-Chain Technology -- 3 Preliminaries -- 3.1 Preliminary Knowledge -- 3.2 System Components -- 4 Solution -- 4.1 Verifiable Complete Data Provenance Query -- 4.2 Efficient Query -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Results and Discussion -- 6 Conclusion -- References -- Enabling Traffic-Differentiated Load Balancing for Datacenter Networks -- 1 Introduction -- 2 Motivation -- 3 TDLB Design -- 3.1 Design Overview -- 3.2 Design Rationale -- 3.3 Design Detail -- 4 Evaluation -- 4.1 Performce Under Symmetric Topology -- 4.2 Performance Under Asymmetric Topology -- 5 Related Work -- 6 Conclusion -- References -- Deep Reinforcement Learning Based Load Balancing for Heterogeneous Traffic in Datacenter Networks -- 1 Introduction -- 2 Background and Motivation -- 2.1 Background -- 2.2 Motivation -- 3 DRLB Design.
3.1 Design Overview -- 3.2 Design Detail -- 4 Experimental Results and Evaluation -- 4.1 Experimental Setup -- 4.2 Evaluation of Experimental Results -- 5 Related Works -- 6 Conclusion -- References -- Adaptive Routing for Datacenter Networks Using Ant Colony Optimization -- 1 Inroduction -- 2 Background and Motivation -- 2.1 Background -- 2.2 Motivation -- 2.3 Summary -- 3 RACO Design -- 3.1 RACO Overview -- 3.2 Statistics Module -- 3.3 Calculation Module -- 3.4 Algorithm Pseudocode -- 4 Performance Evaluation -- 4.1 Experimental Setup -- 4.2 Performance Under Symmetrical Topology -- 4.3 Performance Under Asymmetric Topology -- 5 Related Work -- 6 Conclusion -- References -- MPC: A Novel Internal Clustering Validity Index Based on Midpoint-Involved Distance -- 1 Introduction -- 2 MPdist: A Midpoint-Involved Distance -- 3 MPC: An Internal CVI Based on MPdist -- 3.1 The Definition of MPC -- 3.2 The Analysis of Time Complexity -- 4 Experimental Results -- 4.1 Description of the Test Datasets -- 4.2 Performances of CVIs on Evaluating the Results of the AHC -- 4.3 Performances of CVIs on Evaluating the Results of the K-Means -- 4.4 Performance Evaluation by Real Large-Scale Datasets -- 5 Conclusion and Future Works -- References -- HAECN: Hierarchical Automatic ECN Tuning with Ultra-Low Overhead in Datacenter Networks -- 1 Introduction -- 2 Motivation -- 3 HAECN Design -- 3.1 HAECN Overview -- 3.2 Design Details of HAECN -- 3.3 Reinforcement Learning Agent in HAECN -- 3.4 Learning Algorithm in HAECN -- 4 Implementation -- 5 Evaluation -- 5.1 Performance Under Bursty Networks -- 5.2 Performance Under Gradual Stabilization Network Environments -- 6 Related Work -- 7 Conclusion -- References -- An Android Malware Detection Method Based on Metapath Aggregated Graph Neural Network -- 1 Introduction -- 2 Related Work -- 3 Our Approach.
3.1 Feature Engineering.
Record Nr. UNINA-9910842285903321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part II
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part II
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (524 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-01-X
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part II -- LearnedSync: A Learning-Based Sync Optimization for Cloud Storage -- 1 Introduction -- 2 Background and Motivation -- 2.1 Cloud Sync -- 2.2 Workload Characteristics and Environmental Conditions -- 2.3 Motivation -- 3 The Design of LearnedSync -- 3.1 System Overview -- 3.2 State Monitor -- 3.3 Sync Method Selector -- 3.4 Sync Record Directory -- 3.5 The Training Process for LearnedSync -- 4 Performance Evaluations -- 4.1 Prototype Implementation -- 4.2 Experimental Setup and Methodology -- 4.3 Performance Results and Analysis -- 5 Related Work -- 6 Conclusion -- References -- Optimizing CSR-Based SpMV on a New MIMD Architecture Pezy-SC3s -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Row Granularity Matrix Partition -- 3.2 Workload Balance Within Matrix Chunks -- 3.3 Vectorization -- 4 Experimental Results and Evaluation -- 4.1 Preprocessing Overhead -- 4.2 Floating-Point Performance -- 4.3 Bandwidth Utilization -- 4.4 SpeedUp -- 5 Conclusion -- References -- Intrusion Detection Method for Networked Vehicles Based on Data-Enhanced DBN -- 1 Introduction -- 2 Intrusion Detection Model Based on Data Augmentation -- 2.1 Data Processing and Augmentation -- 2.2 Improved DBN Model -- 3 Experimental Design and Result Analysis -- 3.1 Dataset Labels -- 3.2 Experiments and Analysis -- 4 Conclusion -- References -- A Multi-source Domain Adaption Approach to Minority Disk Failure Prediction -- 1 Introduction -- 2 Related Work -- 3 Motivation -- 3.1 Problem Statement -- 3.2 Generalization Error Bound Analysis -- 4 Method -- 4.1 Overview of DiskDA -- 4.2 Domain Invariant Representation Learning -- 4.3 Confidence-Based Sample Selection -- 5 Experiment -- 5.1 Methodology -- 5.2 Experimental Results -- 5.3 Sensitivity Study -- 6 Conclusion -- 7 Appendix -- References.
Sequenced Quantization RNN Offloading for Dependency Task in Mobile Edge Computing -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formula -- 3.1 System Model -- 3.2 Problem Formula -- 4 Sequenced Quantization Based on RNN Offloading Algorithm -- 4.1 Generation of Unload Actions -- 4.2 Sequenced Quantization Offloading Actions -- 4.3 Experience Pool Recycling -- 5 Simulation Experiments and Result Analysis -- 5.1 Experimental Setup -- 5.2 Algorithm Parameter Selection -- 5.3 Result Analysis -- 5.4 The Impact of the Number of Subtasks on the SQ-RNN Algorithm -- 6 Conclusion -- References -- KylinArm: An Arm Gesture Recognition System for Mobile Devices -- 1 Introduction -- 2 Related Work -- 2.1 IMU-Based Gesture Recognition -- 2.2 Algorithms for Sensor-Based Gesture Recognition -- 3 Design and Implementation of KylinArm -- 3.1 Framework Design of KylinArm -- 3.2 Gesture-Controlled Command Set -- 3.3 Classification Module: Dual-Branch 1D-CNN Classifier -- 3.4 Arm Gesture Recognition -- 3.5 CORAL-Based Generalization Optimization -- 4 Evaluation -- 4.1 DataSet Description -- 4.2 Performance of Dual-Branch 1D-CNN Model -- 4.3 Performance of Inference Modules -- 4.4 Performance of CORALREVERSE -- 4.5 Real-World Evaluation of KylinArm -- 5 Summary -- References -- FCSO: Source Code Summarization by Fusing Multiple Code Features and Ensuring Self-consistency Output -- 1 Introduction -- 2 Related Work -- 3 The Architecture of Approach -- 3.1 Data Preprocessing -- 3.2 Feature Extraction -- 3.3 Feature Fusion -- 3.4 Self-consistency Output -- 4 Experiment -- 4.1 Experiment Settings -- 4.2 Comparison Experiment and Ablation Experiment -- 4.3 Code Summarization Examples -- 5 Conclusion -- References -- Graph Structure Learning-Based Compression Method for Convolutional Neural Networks -- 1 Introduction -- 2 Related Work.
2.1 Compression of CNNs -- 2.2 Graph Structure Learning -- 3 The Proposed Method -- 3.1 Overview -- 3.2 Graph Construction -- 3.3 Graph Structure Learner with Dual-Branch -- 3.4 Pruning Filters -- 4 Experiments -- 4.1 Pruned VGG-16 -- 4.2 Pruned ResNet-101 -- 5 Conclusions and Future Works -- References -- Reliability-Aware VNF Provisioning in Homogeneous and Heterogeneous Multi-access Edge Computing -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Network Model -- 3.2 Task Model -- 3.3 Cost Model -- 4 Homogeneous MEC Scenario -- 4.1 Reliability Model -- 4.2 Problem Formulation -- 4.3 Local Ratio Based Algorithm -- 4.4 Algorithm Analysis -- 5 Heterogeneous MEC Scenario -- 5.1 Reliability Model -- 5.2 Problem Formulation -- 5.3 Benefit-Cost Ratio Preference Algorithm -- 5.4 Algorithm Analysis -- 6 Simulation Evaluation -- 6.1 Experimental Parameters -- 6.2 Impact of Number of Tasks -- 6.3 Impact of Number of Servers -- 7 Conclusion -- References -- Approximate Query Processing Based on Approximate Materialized View -- 1 Introduction -- 2 Overview -- 2.1 Problem Definition -- 2.2 Data Aggregation -- 3 Reuse of Approximate Materialized View -- 3.1 Aggregation and Precomputing -- 3.2 Approximate Interval Recognition -- 3.3 Analysis for Twice Approximations -- 4 Experimental Results -- 4.1 Experimental Setup -- 4.2 Accuracy -- 4.3 The Summary of Experimental Results -- 5 Influence of Parameters -- 5.1 Sample Size -- 5.2 Number of Partition Points -- 5.3 The Summary of Parameter Setting -- 6 Related Work -- 7 Conclusion -- References -- Schema Integration on Massive Data Sources -- 1 Introduction -- 2 Preliminary -- 2.1 Knowledge Base -- 2.2 Distance Function -- 2.3 Edit Distance and Q-Gram -- 2.4 Problem Definition -- 3 Overview -- 3.1 Initialization -- 3.2 Batch Integration -- 3.3 Incremental Integration -- 3.4 Verification.
4 Join Schema Integration -- 4.1 ED Join -- 4.2 Semantic Join -- 5 Batch Integration -- 5.1 Flow of Batch Integration -- 5.2 Resolving -- 6 Experiments -- 6.1 Experimental Settings -- 6.2 Accuracy -- 6.3 Efficiency -- 7 Related Work -- 8 Conclusions and Future Work -- References -- A Hybrid Few-Shot Learning Based Intrusion Detection Method for Internet of Vehicles -- 1 Introduction -- 2 Design of Intrusion Detection Model for IoV -- 2.1 Data Pre-processing -- 2.2 Feature Extraction -- 3 Meta-learning Training Network Process -- 4 Simulation and Results Evaluation -- 4.1 Metrics -- 4.2 Experimental Results and Comparative Analysis -- 5 Conclusion -- References -- Noise-Robust Gaussian Distribution Based Imbalanced Oversampling -- 1 Introduction -- 2 Related Work -- 2.1 Resampling Methods -- 2.2 Gaussian Distribution Based Oversampling (GDO) -- 3 Proposed Method -- 3.1 Analysis of the GDO Algorithm -- 3.2 Local Information of Samples -- 3.3 Estimation of Weight for Minority Class Sample Selection -- 3.4 Probabilistic Seed Sample Selection and Time Complexity -- 4 Experiments and Analysis -- 4.1 Experimental Settings -- 4.2 Experimental Results and Analysis -- 5 Conclusion -- References -- LAST: An Efficient In-place Static Binary Translator for RISC Architectures -- 1 Introduction -- 2 Background -- 2.1 Static Binary Translation -- 2.2 Address Relocation Problem in SBT -- 2.3 Performance Overhead in SBT -- 3 Design -- 3.1 Overview -- 3.2 In-place Instruction Translation Design -- 3.3 System Call Design -- 3.4 ISA-Level Support -- 4 Implementation -- 4.1 Register Mapping -- 4.2 In-place Instruction Translation -- 4.3 System Call Handling -- 4.4 Delay Slot -- 5 Evaluation -- 5.1 Evaluation Setup -- 5.2 Performance -- 5.3 Translation vs. Compilation -- 6 Conclusion -- References.
Personalized Privacy Risk Assessment Based on Deep Neural Network for Image Sharing on Social Networks -- 1 Introduction -- 2 Related Work -- 3 The Framework of PPRAS -- 3.1 Problem Description and Definition -- 3.2 The Framework of PPRAS -- 4 Experimental Results and Evaluation -- 4.1 Dataset -- 4.2 Evaluation Metrics and Baseline Methods -- 4.3 Train the Neural Network -- 4.4 Comparison of PPRAS and Two Variants in Three Performance Metrics -- 4.5 Comparison of PPRAS and Two Variants in Time Utility -- 4.6 Utility Analysis -- 4.7 Parametric Analysis -- 5 Conclusions -- References -- A Pipelined AES and SM4 Hardware Implementation for Multi-tasking Virtualized Environments -- 1 Introduction -- 2 Background -- 2.1 Overview of AES and SM4 -- 2.2 Related Works and Discussion -- 3 Proposed AES and SM4 Architecture -- 3.1 AES Encryption and Decryption Architecture -- 3.2 SM4 Encryption/Decryption Architecture -- 4 A Prototype Cryptosystem for Virtualization -- 4.1 Prototype Cryptosystem Architecture -- 4.2 Programming Sequence of the Prototype Cryptosystem -- 5 Experimental Results -- 5.1 Circuit Performance Comparison -- 5.2 Multi-guest Efficiency Improvement -- 6 Discussion -- 7 Conclusion -- References -- Blockchain-Assisted Privacy-Preserving Public Auditing Scheme for Cloud Storage Systems -- 1 Introduction -- 2 Related Work -- 3 Technical Preliminaries -- 3.1 Bilinear Pairing and Hardness Problem -- 3.2 Homomorphic Hash Function -- 3.3 System Model -- 3.4 Adversary Model and Design Goals -- 4 Our Protocol -- 5 Evaluation of the Proposed Mechanism -- 5.1 Correctness -- 5.2 Security Analysis -- 6 Comprehensive Performance Evaluation -- 6.1 Comparison of the Computation Costs -- 6.2 Comparison of the Communication Overhead -- 6.3 On-Chain Consumption Evaluation -- 7 Conclusions and Future Work -- References.
MANet: An Architecture Adaptive Method for Sparse Matrix Format Selection.
Record Nr. UNINA-9910842292303321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VII
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VII
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (375 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-62-1
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part VII -- An Efficient Scheduling Algorithm for Multi-mode Tasks on Near-Data Processing SSDs -- 1 Introduction -- 2 Background and Motivation -- 2.1 Task Processing of NDP-Based SSDs -- 2.2 Motivation Example -- 3 Problem Definition -- 4 Multi-mode Task Scheduling -- 4.1 Problem Analysis -- 4.2 MMTS Algorithm -- 5 Evaluation -- 5.1 Experimental Setups -- 5.2 Effect on Performance -- 5.3 Resource Utilization -- 6 Conclusion -- References -- HR-kESP: A Heuristic Algorithm for Robustness-Oriented k Edge Server Placement -- 1 Introduction -- 2 Problem and Model Formulation -- 2.1 Problem Statement -- 2.2 User Coverage -- 2.3 Network Robustness -- 2.4 Robustness-Oriented kESP Optimization Model -- 3 The Heuristic Algorithm for kESP -- 3.1 Selection of Initial Base Station -- 3.2 Gradual Generation of Server Deployment Scheme -- 3.3 HR-kESP Algorithm -- 4 Experimental Evaluation -- 4.1 Experimental Settings -- 4.2 Comparison on Different Numbers of Base Stations -- 4.3 Comparison on Different Server Budgets -- 4.4 Comparison on Different Numbers of Users -- 5 Related Work -- 6 Conclusion and Future Work -- References -- A Hybrid Kernel Pruning Approach for Efficient and Accurate CNNs -- 1 Introduction -- 2 Functional-Similarity-Based Kernel Pruning Algorithm -- 2.1 Distance-Based Kernel Pruning Algorithm -- 2.2 Hybrid Kernel Pruning Algorithm -- 3 Experimental Method -- 4 Experimental Results -- 4.1 Experimental Results for the Distance-Based Pruning Algorithm -- 4.2 Experimental Results for the Hybrid Pruning Algorithm -- 5 Analysis -- 6 Related Work -- 7 Conclusion -- References -- A Collaborative Migration Algorithm for Edge Services Based on Evolutionary Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 Service Migration Model -- 3.1 Scene Description.
3.2 Service Migration Model Construction -- 3.3 Problem Formulation -- 4 Service Migration Algorithm Based on Evolutionary Reinforcement Learning -- 4.1 MDP Model -- 4.2 DEDRL-Based Service Migration Process -- 5 Experiment and Results -- 5.1 Experimental Environment and Parameter Settings -- 5.2 Analysis of Results -- 6 Conclusion -- References -- A Graph Generation Network with Privacy Preserving Capabilities -- 1 Introduction -- 2 Related Works -- 3 Preliminaries -- 3.1 Generative Adversarial Networks -- 3.2 Differential Privacy -- 4 RDPGGAN -- 5 Experimental Evaluations -- 6 Conclusion -- References -- Clustered Federated Learning Framework with Acceleration Based on Data Similarity -- 1 Introduction -- 2 Preliminaries -- 2.1 Federated Learning -- 2.2 Locality-Sensitive Hashing -- 3 Design of Our FedCSA Framework -- 3.1 Intra-group Acceleration -- 3.2 Group Aggregation -- 3.3 Privacy Analysis -- 4 Experiment and Result -- 4.1 Experiment Settings -- 4.2 Result and Analysis -- 5 Conclusion -- References -- An Anonymous Authentication Scheme with Low Overhead for Cross-Domain IoT -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 MHT and MHG -- 3.2 Zero-Knowledge Proof -- 4 Proposed Scheme -- 4.1 System Overview -- 4.2 Threat Model -- 5 Design Detail -- 5.1 Initialization Phase -- 5.2 Device Registration -- 5.3 Intra-domain Authentication -- 5.4 Cross-Domain Authentication -- 5.5 Key Agreement -- 5.6 Data Correctness Verification -- 6 Theoretical Analysis of CALA -- 7 Experiment and Evaluation -- 7.1 Experimental Settings -- 7.2 Performance Analysis -- 8 Conclusion -- References -- UAV-Assisted Data Collection and Transmission Using Petal Algorithm in Wireless Sensor Networks -- 1 Introduction -- 2 Mathematical Model for the UAV-Assisted Data Collection in WSNs -- 2.1 WSN Model and Fundamental Concepts.
2.2 Mathematical Model of the UAV-Assisted Data Collection -- 3 Petal Algorithm for the UAV-Assisted Data Collection -- 3.1 Analysis of the UAV-Assisted Data Collection in WSNs -- 3.2 Petal Algorithm for the UAV-Assisted Data Collection in WSNs -- 4 Experimental Simulation -- 4.1 Test Problem and Parameter Description -- 4.2 Comparative Analysis of Experimental Results -- 5 Conclusion -- References -- DeletePop: A DLT Execution Time Predictor Based on Comprehensive Modeling -- 1 Introduction -- 2 Background and Current Challenges -- 2.1 Distributed Strategy for Deep Learning Training -- 2.2 Job Scheduling Simulator -- 2.3 Major Challenges of Modeling and Simulation of Deep Learning Training Job -- 3 Cost Model Design -- 3.1 Modeling of the DLT Process -- 3.2 The Predicting Algorithm of the Cost Model -- 4 Simulator Design and Implementation -- 5 Evaluation -- 5.1 Prediction of Step Time (The Atomic Unit) -- 5.2 Prediction of Epoch Time -- 5.3 Performance Simulation Experiment -- 6 Related Work -- 7 Conclusion -- References -- CFChain: A Crowdfunding Platform that Supports Identity Authentication, Privacy Protection, and Efficient Audit -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Bilinear Groups -- 3.2 BLS Aggregate Signature ch10BLS -- 3.3 Commitment and Non-interactive Zero-Knowledge Proofs -- 3.4 Distributed Identities -- 4 CFChain Overview -- 4.1 System Model -- 4.2 Threat Model and Design Goals -- 4.3 System Operations -- 5 Security Definitions -- 6 The CFChain Scheme -- 7 Security Analyse -- 8 Performance Evaluation -- 8.1 Functionality Comparison -- 8.2 Implementation -- 8.3 Performance -- 9 Conclusion -- References -- TBAF: A Two-Stage Biometric-Assisted Authentication Framework in Edge-Integrated UAV Delivery System -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Physical Unclonable Function.
3.2 Blakley Secret Sharing -- 3.3 Zero Knowledge Proofs -- 3.4 Cancelable Biometric Template Protection -- 4 The Proposed Scheme -- 4.1 System Overview -- 4.2 Three-Dimensional Space-Based Secret Sharing -- 4.3 Registration -- 4.4 Authentication and Key Negotiation -- 5 Evaluation -- 5.1 Formal Security Analysis Using BAN Logic -- 5.2 Informal Security Analysis -- 5.3 Efficiency Analysis -- 6 Conclusion and Future Work -- References -- Attention Enhanced Package Pick-Up Time Prediction via Heterogeneous Behavior Modeling -- 1 Introduction -- 2 Overview -- 2.1 Preliminaries -- 2.2 Problem Formulation -- 3 Model Design -- 3.1 Overview -- 3.2 Pre-trained Stay Time Prediction -- 3.3 Attention Enhanced Route Estimation -- 3.4 Training and Prediction -- 4 Evaluation -- 4.1 Dataset Description -- 4.2 Experimental Settings -- 4.3 Metrics -- 4.4 Baselines -- 4.5 Main Performance -- 4.6 Ablation Studies of the Pre-trained Module -- 4.7 Real-World Deployment -- 5 Discussion -- 5.1 Lessons Learned -- 5.2 Limitations and Future Works -- 6 Related Works -- 6.1 Estimate Time of Arrival Applications -- 6.2 Route Prediction in Transportation and Logistics Systems -- 6.3 Uniqueness of Our Work -- 7 Conclusion -- References -- Optimizing Pointwise Convolutions on Multi-core DSPs -- 1 Introduction and Related Work -- 2 Backgound -- 2.1 Pointwise Convolution -- 2.2 Architecture of FT-M7032 Heterogeneous Processors -- 3 Parallel Direct Implementation -- 3.1 Overview of Our Implementation -- 3.2 Multi-level Parallel Forward Propagation Algorithm -- 3.3 Multi-level Parallel Algorithms for Backward Propagation and Weight Gradient Update Propagation -- 4 Performance Evaluation -- 4.1 Experiment Setup -- 4.2 Performance -- 5 Conclusions and Future Work -- References -- Detecting SDCs in GPGPUs Through Efficient Partial Thread Redundancy -- 1 Introduction -- 2 Background.
2.1 GPGPUs Architecture and Programming Model -- 2.2 Fault Model -- 3 Thread-Level SDC Proneness Analysis -- 3.1 Vulnerability Identification -- 3.2 Intra-thread Error Propagation Analysis -- 3.3 Inter-thread Error Propagation Analysis -- 4 Partial Thread Protection Framework -- 4.1 Thread SDC Vulnerability Profiling -- 4.2 Partial Thread Redundancy -- 5 Experiment Methodology -- 6 Evaluation -- 6.1 SDC Probability Prediction Accuracy -- 6.2 Overhead -- 7 Related Work -- 8 Conclusion -- References -- FDRShare: A Fully Decentralized and Redactable EHRs Sharing Scheme with Constant-Size Ciphertexts -- 1 Introduction -- 1.1 Our Results and Contributions -- 2 Related Work -- 3 Preliminaries -- 3.1 Bilinear Pairing -- 3.2 DDH Assumption -- 3.3 Pseudo-Random Function -- 4 System Definition -- 4.1 Threat Model -- 4.2 System Model -- 5 Building Block -- 6 Detail Construction of FDRShare -- 7 Conclusion -- References -- An Efficient Fault Tolerance Strategy for Multi-task MapReduce Models Using Coded Distributed Computing -- 1 Introduction -- 2 Related Work -- 3 Preliminaries and Problem Formulation -- 3.1 The MapReduce Framework -- 3.2 Coded Distributed Computing -- 3.3 Problem Formulation -- 4 Fault Tolerance for Multiple Reduce Tasks Using Coded Distributed Computing -- 4.1 Preparation Stage -- 4.2 Map Stage -- 4.3 Shuffle Stage -- 4.4 Reduce Stage -- 5 Experiments -- 5.1 Experiment Setups -- 5.2 Experimental Results -- 6 Summary -- References -- Key-Based Transaction Reordering: An Optimized Approach for Concurrency Control in Hyperledger Fabric -- 1 Introduction -- 2 Related Work -- 2.1 Endorsement Phase Optimization -- 2.2 Ordering Phase Optimization -- 2.3 Validation Phase Optimization -- 3 Problem Definition -- 3.1 Types of Transaction Conflicts -- 3.2 Problem Formulation -- 4 System Design -- 4.1 System Model -- 4.2 Algorithm Design.
5 Experimental Evaluation.
Record Nr. UNINA-9910842293003321
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. Federico II
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VII
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VII
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (375 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-62-1
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part VII -- An Efficient Scheduling Algorithm for Multi-mode Tasks on Near-Data Processing SSDs -- 1 Introduction -- 2 Background and Motivation -- 2.1 Task Processing of NDP-Based SSDs -- 2.2 Motivation Example -- 3 Problem Definition -- 4 Multi-mode Task Scheduling -- 4.1 Problem Analysis -- 4.2 MMTS Algorithm -- 5 Evaluation -- 5.1 Experimental Setups -- 5.2 Effect on Performance -- 5.3 Resource Utilization -- 6 Conclusion -- References -- HR-kESP: A Heuristic Algorithm for Robustness-Oriented k Edge Server Placement -- 1 Introduction -- 2 Problem and Model Formulation -- 2.1 Problem Statement -- 2.2 User Coverage -- 2.3 Network Robustness -- 2.4 Robustness-Oriented kESP Optimization Model -- 3 The Heuristic Algorithm for kESP -- 3.1 Selection of Initial Base Station -- 3.2 Gradual Generation of Server Deployment Scheme -- 3.3 HR-kESP Algorithm -- 4 Experimental Evaluation -- 4.1 Experimental Settings -- 4.2 Comparison on Different Numbers of Base Stations -- 4.3 Comparison on Different Server Budgets -- 4.4 Comparison on Different Numbers of Users -- 5 Related Work -- 6 Conclusion and Future Work -- References -- A Hybrid Kernel Pruning Approach for Efficient and Accurate CNNs -- 1 Introduction -- 2 Functional-Similarity-Based Kernel Pruning Algorithm -- 2.1 Distance-Based Kernel Pruning Algorithm -- 2.2 Hybrid Kernel Pruning Algorithm -- 3 Experimental Method -- 4 Experimental Results -- 4.1 Experimental Results for the Distance-Based Pruning Algorithm -- 4.2 Experimental Results for the Hybrid Pruning Algorithm -- 5 Analysis -- 6 Related Work -- 7 Conclusion -- References -- A Collaborative Migration Algorithm for Edge Services Based on Evolutionary Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 Service Migration Model -- 3.1 Scene Description.
3.2 Service Migration Model Construction -- 3.3 Problem Formulation -- 4 Service Migration Algorithm Based on Evolutionary Reinforcement Learning -- 4.1 MDP Model -- 4.2 DEDRL-Based Service Migration Process -- 5 Experiment and Results -- 5.1 Experimental Environment and Parameter Settings -- 5.2 Analysis of Results -- 6 Conclusion -- References -- A Graph Generation Network with Privacy Preserving Capabilities -- 1 Introduction -- 2 Related Works -- 3 Preliminaries -- 3.1 Generative Adversarial Networks -- 3.2 Differential Privacy -- 4 RDPGGAN -- 5 Experimental Evaluations -- 6 Conclusion -- References -- Clustered Federated Learning Framework with Acceleration Based on Data Similarity -- 1 Introduction -- 2 Preliminaries -- 2.1 Federated Learning -- 2.2 Locality-Sensitive Hashing -- 3 Design of Our FedCSA Framework -- 3.1 Intra-group Acceleration -- 3.2 Group Aggregation -- 3.3 Privacy Analysis -- 4 Experiment and Result -- 4.1 Experiment Settings -- 4.2 Result and Analysis -- 5 Conclusion -- References -- An Anonymous Authentication Scheme with Low Overhead for Cross-Domain IoT -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 MHT and MHG -- 3.2 Zero-Knowledge Proof -- 4 Proposed Scheme -- 4.1 System Overview -- 4.2 Threat Model -- 5 Design Detail -- 5.1 Initialization Phase -- 5.2 Device Registration -- 5.3 Intra-domain Authentication -- 5.4 Cross-Domain Authentication -- 5.5 Key Agreement -- 5.6 Data Correctness Verification -- 6 Theoretical Analysis of CALA -- 7 Experiment and Evaluation -- 7.1 Experimental Settings -- 7.2 Performance Analysis -- 8 Conclusion -- References -- UAV-Assisted Data Collection and Transmission Using Petal Algorithm in Wireless Sensor Networks -- 1 Introduction -- 2 Mathematical Model for the UAV-Assisted Data Collection in WSNs -- 2.1 WSN Model and Fundamental Concepts.
2.2 Mathematical Model of the UAV-Assisted Data Collection -- 3 Petal Algorithm for the UAV-Assisted Data Collection -- 3.1 Analysis of the UAV-Assisted Data Collection in WSNs -- 3.2 Petal Algorithm for the UAV-Assisted Data Collection in WSNs -- 4 Experimental Simulation -- 4.1 Test Problem and Parameter Description -- 4.2 Comparative Analysis of Experimental Results -- 5 Conclusion -- References -- DeletePop: A DLT Execution Time Predictor Based on Comprehensive Modeling -- 1 Introduction -- 2 Background and Current Challenges -- 2.1 Distributed Strategy for Deep Learning Training -- 2.2 Job Scheduling Simulator -- 2.3 Major Challenges of Modeling and Simulation of Deep Learning Training Job -- 3 Cost Model Design -- 3.1 Modeling of the DLT Process -- 3.2 The Predicting Algorithm of the Cost Model -- 4 Simulator Design and Implementation -- 5 Evaluation -- 5.1 Prediction of Step Time (The Atomic Unit) -- 5.2 Prediction of Epoch Time -- 5.3 Performance Simulation Experiment -- 6 Related Work -- 7 Conclusion -- References -- CFChain: A Crowdfunding Platform that Supports Identity Authentication, Privacy Protection, and Efficient Audit -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Bilinear Groups -- 3.2 BLS Aggregate Signature ch10BLS -- 3.3 Commitment and Non-interactive Zero-Knowledge Proofs -- 3.4 Distributed Identities -- 4 CFChain Overview -- 4.1 System Model -- 4.2 Threat Model and Design Goals -- 4.3 System Operations -- 5 Security Definitions -- 6 The CFChain Scheme -- 7 Security Analyse -- 8 Performance Evaluation -- 8.1 Functionality Comparison -- 8.2 Implementation -- 8.3 Performance -- 9 Conclusion -- References -- TBAF: A Two-Stage Biometric-Assisted Authentication Framework in Edge-Integrated UAV Delivery System -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 3.1 Physical Unclonable Function.
3.2 Blakley Secret Sharing -- 3.3 Zero Knowledge Proofs -- 3.4 Cancelable Biometric Template Protection -- 4 The Proposed Scheme -- 4.1 System Overview -- 4.2 Three-Dimensional Space-Based Secret Sharing -- 4.3 Registration -- 4.4 Authentication and Key Negotiation -- 5 Evaluation -- 5.1 Formal Security Analysis Using BAN Logic -- 5.2 Informal Security Analysis -- 5.3 Efficiency Analysis -- 6 Conclusion and Future Work -- References -- Attention Enhanced Package Pick-Up Time Prediction via Heterogeneous Behavior Modeling -- 1 Introduction -- 2 Overview -- 2.1 Preliminaries -- 2.2 Problem Formulation -- 3 Model Design -- 3.1 Overview -- 3.2 Pre-trained Stay Time Prediction -- 3.3 Attention Enhanced Route Estimation -- 3.4 Training and Prediction -- 4 Evaluation -- 4.1 Dataset Description -- 4.2 Experimental Settings -- 4.3 Metrics -- 4.4 Baselines -- 4.5 Main Performance -- 4.6 Ablation Studies of the Pre-trained Module -- 4.7 Real-World Deployment -- 5 Discussion -- 5.1 Lessons Learned -- 5.2 Limitations and Future Works -- 6 Related Works -- 6.1 Estimate Time of Arrival Applications -- 6.2 Route Prediction in Transportation and Logistics Systems -- 6.3 Uniqueness of Our Work -- 7 Conclusion -- References -- Optimizing Pointwise Convolutions on Multi-core DSPs -- 1 Introduction and Related Work -- 2 Backgound -- 2.1 Pointwise Convolution -- 2.2 Architecture of FT-M7032 Heterogeneous Processors -- 3 Parallel Direct Implementation -- 3.1 Overview of Our Implementation -- 3.2 Multi-level Parallel Forward Propagation Algorithm -- 3.3 Multi-level Parallel Algorithms for Backward Propagation and Weight Gradient Update Propagation -- 4 Performance Evaluation -- 4.1 Experiment Setup -- 4.2 Performance -- 5 Conclusions and Future Work -- References -- Detecting SDCs in GPGPUs Through Efficient Partial Thread Redundancy -- 1 Introduction -- 2 Background.
2.1 GPGPUs Architecture and Programming Model -- 2.2 Fault Model -- 3 Thread-Level SDC Proneness Analysis -- 3.1 Vulnerability Identification -- 3.2 Intra-thread Error Propagation Analysis -- 3.3 Inter-thread Error Propagation Analysis -- 4 Partial Thread Protection Framework -- 4.1 Thread SDC Vulnerability Profiling -- 4.2 Partial Thread Redundancy -- 5 Experiment Methodology -- 6 Evaluation -- 6.1 SDC Probability Prediction Accuracy -- 6.2 Overhead -- 7 Related Work -- 8 Conclusion -- References -- FDRShare: A Fully Decentralized and Redactable EHRs Sharing Scheme with Constant-Size Ciphertexts -- 1 Introduction -- 1.1 Our Results and Contributions -- 2 Related Work -- 3 Preliminaries -- 3.1 Bilinear Pairing -- 3.2 DDH Assumption -- 3.3 Pseudo-Random Function -- 4 System Definition -- 4.1 Threat Model -- 4.2 System Model -- 5 Building Block -- 6 Detail Construction of FDRShare -- 7 Conclusion -- References -- An Efficient Fault Tolerance Strategy for Multi-task MapReduce Models Using Coded Distributed Computing -- 1 Introduction -- 2 Related Work -- 3 Preliminaries and Problem Formulation -- 3.1 The MapReduce Framework -- 3.2 Coded Distributed Computing -- 3.3 Problem Formulation -- 4 Fault Tolerance for Multiple Reduce Tasks Using Coded Distributed Computing -- 4.1 Preparation Stage -- 4.2 Map Stage -- 4.3 Shuffle Stage -- 4.4 Reduce Stage -- 5 Experiments -- 5.1 Experiment Setups -- 5.2 Experimental Results -- 6 Summary -- References -- Key-Based Transaction Reordering: An Optimized Approach for Concurrency Control in Hyperledger Fabric -- 1 Introduction -- 2 Related Work -- 2.1 Endorsement Phase Optimization -- 2.2 Ordering Phase Optimization -- 2.3 Validation Phase Optimization -- 3 Problem Definition -- 3.1 Types of Transaction Conflicts -- 3.2 Problem Formulation -- 4 System Design -- 4.1 System Model -- 4.2 Algorithm Design.
5 Experimental Evaluation.
Record Nr. UNISA-996587859703316
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. di Salerno
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part IV
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part IV
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (508 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-59-1
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part IV -- CSDSE: Apply Cooperative Search to Solve the Exploration-Exploitation Dilemma of Design Space Exploration -- 1 Introduction -- 2 Background and Motivation -- 2.1 Design Space of Deep Neural Network Accelerator -- 2.2 Design Space Exploration Methods -- 2.3 Challenge of Exploration-Exploitation Dilemma -- 3 CSDSE Method -- 3.1 Profile of CSDSE -- 3.2 Definition of RL Framework -- 3.3 Cooperative Search and Heterogeneous Agents -- 3.4 Weighted Compact Buffer -- 4 Experiments and Analyses -- 4.1 Experiment Setup -- 4.2 Experiment Result and Analysis -- 5 Related Works -- 6 Conclusion -- References -- Differential Privacy in Federated Dynamic Gradient Clipping Based on Gradient Norm -- 1 Introduction -- 2 Related Work -- 3 Preliminary -- 3.1 Threat Model -- 3.2 Gradient Cropping -- 4 Differential Privacy in Federated Dynamic Gradient Clipping Based on Gradient Norm -- 4.1 Overall Framework -- 4.2 Relationship Between Gradient Variation and Clipped Threshold -- 4.3 Gradient Attenuation Clipped Threshold Based on Gradient Norm -- 4.4 Convergence Analysis -- 4.5 Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Setup -- 5.2 Performance of Privacy Protection -- 5.3 Global Model Availability -- 5.4 Universality of Clipped Thresholds -- 6 Conclusions -- References -- SolGPT: A GPT-Based Static Vulnerability Detection Model for Enhancing Smart Contract Security -- 1 Introduction -- 2 Problem Formulation -- 3 Related Work -- 3.1 Deep Learning-Based Approaches for Smart Contract Vulnerability Detection -- 3.2 GPT Models -- 4 SolGPT: Solidity Adapative GPT -- 4.1 SolTokenizer -- 4.2 Solidity Adaptive Pre-training -- 4.3 Vulnerability Detection Fine-Tunning -- 5 Experiment and Analysis -- 5.1 Comparisons with Traditional Approaches -- 5.2 Comparisons with Deep Learning Approaches.
5.3 Ablation Studies -- 6 Conclusion -- References -- Neuron Pruning-Based Federated Learning for Communication-Efficient Distributed Training -- 1 Introduction -- 2 Related Work -- 2.1 FL with Edge Computing -- 2.2 Communication Overhead Reduction in FL -- 3 Algorithm Description -- 3.1 Basic Model Selection -- 3.2 Model Pruning -- 4 Simulation Results and Analysis -- 4.1 Experimental Setting -- 4.2 Performance Analysis -- 4.3 Network Traffic Classification -- 5 Conclusion and Future Work -- References -- A Novel Network Topology Sensing Method for Network Security Situation Awareness -- 1 Introduction -- 2 Preliminaries and Related Works -- 2.1 Network Security Situation Awareness (NSSA) -- 2.2 Group Decision Making (GDM) -- 3 Network Topology Sensing Method for NSSA -- 3.1 Overview -- 3.2 Agent Model -- 3.3 Group Decision Making Based on Trusted Agents -- 4 Experimental Studies and Analysis -- 4.1 Prerequisites and Problem Descriptions -- 4.2 Experimental Process and Numerical Example -- 4.3 Comparisons and Discussions -- 5 Conclusions -- References -- Distributed Task Offloading Method Based on Federated Reinforcement Learning in Vehicular Networks with Incomplete Information -- 1 Introduction -- 2 Related Work -- 3 System Model -- 3.1 Local Computing -- 3.2 Edge Computing -- 4 Problem Formulation -- 5 Algorithm Design -- 5.1 Deep Q-Network -- 5.2 State Space -- 5.3 Action Space -- 5.4 Reward Function -- 5.5 Communication Mask -- 5.6 FDQN Algorithm -- 6 Performance Evaluation -- 6.1 Simulation Setup -- 6.2 Evaluation Results -- 7 Conclusion -- References -- Carbon Trading Based on Consortium Chain: Building, Modeling, and Analysis -- 1 Introduction -- 2 DC-Chain for Carbon Trading -- 2.1 Role Assignment -- 2.2 DC-Chain Construction and Smart Contract Design -- 3 Modeling Blockchain Performance -- 4 Analytical and Experimental Results.
4.1 Transaction Confirmation Time -- 4.2 Block Size -- 5 Conclusion -- References -- CAST: An Intricate-Scene Aware Adaptive Bitrate Approach for Video Streaming via Parallel Training -- 1 Introduction -- 2 Background and Motivation -- 2.1 ABR Algorithms -- 2.2 Scene Complexity -- 3 Design -- 3.1 CAST Overview -- 3.2 Neural Network Structure -- 3.3 Implementation -- 4 Evaluation -- 4.1 Methodology -- 4.2 Comparison with Existing ABR Algorithms -- 4.3 CAST vs. Existing ABR Approaches -- 4.4 Performance Under Different Network Environments -- 4.5 Performance Under Different Player Settings -- 4.6 Subjective Tests from Real Users -- 5 Conclusions -- References -- A Hybrid Active and Passive Cache Method Based on Deep Learning in Edge Computing -- 1 Introduction -- 2 Related Work -- 3 System Model -- 4 Proposed Methodology -- 4.1 Feature Extraction from Four Modalities -- 4.2 Multimodal Prediction Model -- 5 Experiment Setup -- 5.1 Dataset Setup -- 5.2 Evaluating Indicator -- 5.3 Baselines -- 6 Cached Updates -- 7 Evaluation Results -- 7.1 Simulation Setup -- 7.2 Evaluation Results -- 8 Conclusion -- References -- We Will Find You: An Edge-Based Multi-UAV Multi-Recipient Identification Method in Smart Delivery Services -- 1 Introduction -- 2 Related Work -- 3 Framework and Algorithms -- 3.1 Framework -- 3.2 Algorithms -- 4 Evaluation -- 4.1 Case Study -- 4.2 Experiments -- 5 Conclusions and Future Work -- References -- Multi-objective Optimization for Joint Handover Decision and Computation Offloading in Integrated Communications and Computing 6G Networks -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulations -- 3.1 System Model -- 3.2 Communication Model -- 3.3 Computation Model -- 3.4 Problem Formulation -- 4 Joint Handover Decision and Computation Offloading Method Based on MOO -- 4.1 The Flowchart of Mobility Management in CCN.
4.2 Trigger Mechanisms of Handover and Offloading -- 4.3 Joint Decision Method Based on MOO -- 4.4 The Convergence Analysis of JD-MOO Method -- 5 Performance Evaluation -- 5.1 Experiments Setting -- 5.2 Performance of Handover Decision -- 5.3 Performance of Computation Offloading -- 5.4 Analysis of Services Satisfaction Degree -- 5.5 Analysis of Running Time and Convergence -- 6 Conclusion -- References -- A Joint Optimization Scheme in Heterogeneous UAV-Assisted MEC -- 1 Introduction -- 2 Related Work -- 2.1 MEC -- 2.2 UAV Scheduling -- 2.3 Task Allocation -- 3 System Model -- 3.1 Problem Statement -- 3.2 System Overview -- 3.3 UAV Scheduling Model -- 3.4 Task Execution Model -- 3.5 UAV Hovering Model -- 3.6 Problem Formulation -- 4 Proposed Approach -- 4.1 UAV Scheduling Algorithm -- 4.2 Task Allocation Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Settings -- 5.2 Experimental Results -- 6 Conclusion -- References -- Multi-UAV Collaborative Face Recognition for Goods Receiver in Edge-Based Smart Delivery Services -- 1 Introduction -- 2 Framework Design and Workflow -- 2.1 Framework Overview -- 2.2 Framework Execution Process -- 3 Related Technologies and Algorithms -- 3.1 Face Detection -- 3.2 Face Fusion -- 3.3 Face Recognition -- 4 Experiments and Evaluation -- 4.1 Case Studies -- 4.2 Evaluation -- 5 Conclusions -- References -- Distributed Generative Adversarial Networks for Fuzzy Portfolio Optimization -- 1 Introduction -- 2 Generative Adversarial Networks Architecture -- 2.1 Theoretical Background -- 2.2 Architecture -- 2.3 Generator -- 2.4 Discriminator -- 2.5 Adversarial Training -- 3 Generative Adversarial Networks Architecture -- 4 Experimental Results -- 4.1 Experimental Setting -- 4.2 Experimental Results -- 4.3 Experimental Evaluation -- 5 Parallel Fuzzy Portfolio Optimization -- 5.1 Fuzzy Mean-CVaR Portfolio Model.
5.2 Optimization Algorithms -- 5.3 Parallelization and Results -- 6 Conclusion -- References -- A Novel Transaction Processing Model for Sharded Blockchain -- 1 Introduction -- 2 Related Work -- 2.1 Blockchain Sharding Protocols -- 2.2 Transaction Processing -- 3 Proposed Model -- 3.1 Overview of Model -- 3.2 Threat Model -- 4 Transaction Admission Control Algorithm -- 4.1 Problem Formulation -- 4.2 Algorithm Design -- 5 Transaction Verification Design -- 5.1 Verification Based on TSS -- 5.2 Verification Peers Selection -- 5.3 Transaction Dependency Handling -- 6 Experiment and Evaluation -- 6.1 Implementation -- 6.2 Performance Evaluation -- 7 Conclusion -- References -- Computation Offloading Based on Deep Reinforcement Learning for UAV-MEC Network -- 1 Introduction -- 2 System Model and Problem Description -- 2.1 System Model -- 2.2 Communication Model -- 2.3 Task Calculation Model -- 2.4 Problem Formulation -- 3 Proposed Scheme -- 3.1 SDQN Algorithm -- 3.2 Reinforcement Learning Modeling -- 3.3 Training and Testing -- 4 Simulation Results -- 4.1 Simulation Setting -- 4.2 Simulation Results and Discussions -- 5 Conclusion -- References -- Privacy-Aware Scheduling Heuristic Based on Priority in Edge Environment -- 1 Introduction -- 2 Related Work -- 3 Problem Modeling -- 3.1 Application Model -- 3.2 Resource and Processing Time Model -- 4 Method -- 4.1 Task Selection Phase -- 4.2 Processor Selection Phase -- 5 Evaluation -- 5.1 Settings -- 5.2 The Performance Comparison Experiment for Privacy Application -- 5.3 The Performance Comparison Experiment for General Application -- 6 Conclusion -- References -- A Novel Multi-objective Evolutionary Algorithm Hybrid Simulated Annealing Concept for Recommendation Systems -- 1 Introduction -- 2 Related Work -- 3 Problem Definition -- 3.1 Multi-objective Evolutionary Algorithm -- 4 The Proposed Model.
4.1 Objective Functions.
Record Nr. UNISA-996587859603316
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. di Salerno
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part III
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part III
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (524 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9707-98-6
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part III -- Addressing Coupled Constrained Reinforcement Learning via Interative Iteration Design -- 1 Introduction -- 2 Preliminaries -- 2.1 Coupled Constrained Markov Decision Process -- 2.2 Policy Gradient -- 3 InIt for Decoupling High-Dimensional Coupled Action Space -- 3.1 InIt Design -- 3.2 InIt-Based Solution -- 3.3 Extension to High-Dimensional Action Space -- 4 Experiments -- 4.1 Experiment Setups -- 4.2 Baselines -- 4.3 Performance Evaluation -- 5 Conclusion and Future Work -- References -- A Path Planning and Obstacle Avoidance Method for USV Based on Dynamic-Target APF Algorithm in Edge -- 1 Introduction -- 2 Related Works -- 3 System Design and Environment Modeling -- 3.1 System Overview -- 3.2 Environment Modeling -- 4 Algorithm Design -- 4.1 Improved A* Algorithm -- 4.2 Improved APF Algorithm -- 4.3 DTAPF Method -- 4.4 Offset Guidance -- 5 Evaluation -- 5.1 Simulation Environment Configurations -- 5.2 Performance of Algorithms -- 5.3 Verification of System Architecture -- 6 Conclusion -- References -- Node-Disjoint Paths in Balanced Hypercubes with Application to Fault-Tolerant Routing -- 1 Introduction -- 2 Preliminaries -- 3 Node-Disjoint Paths -- 3.1 The Mixed Sequence -- 3.2 Construction of Node-Disjoint Paths -- 4 Simulations and Performance Evaluation -- 5 Conclusion -- References -- Performance Evaluation of Spark, Ray and MPI: A Case Study on Long Read Alignment Algorithm -- 1 Introduction -- 2 Methods -- 2.1 Parallel Algorithm Design -- 2.2 Index Sharing -- 2.3 Sequence Segmentation -- 2.4 Parallel Alignment -- 2.5 Asynchronous Output -- 3 Results -- 3.1 Experimental Setup -- 3.2 Input Data -- 3.3 Results Consistency -- 3.4 Strong Scalability -- 3.5 Weak Scalability and Peak RAM Memory -- 3.6 Efficiency -- 3.7 Throughput -- 3.8 Latency -- 3.9 Load Balance -- 4 Related Work.
5 Discussion -- 6 Conclusion -- References -- Fairness Analysis and Optimization of BBR Congestion Control Algorithm -- 1 Introduction -- 2 Related Work -- 3 Problem Analysis -- 3.1 RTT Fairness in BBR Algorithm -- 4 Algorithm Optimization -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Results on RTT Fairness -- 6 Conclusion -- References -- Accelerating QUIC with AF_XDP -- 1 Introduction -- 2 Background -- 2.1 QUIC Acceleration -- 2.2 Network Stack Optimization -- 2.3 eBPF and AF_XDP -- 2.4 AF_XDP Datapath -- 3 Design and Implementation -- 3.1 Overview -- 3.2 Architecture -- 3.3 Implementation -- 4 Evaluation -- 4.1 Experiment Setup -- 4.2 QUIC Performance -- 4.3 CPU Usage -- 4.4 Network Policy -- 5 Discussion -- 6 Conclusion -- References -- Segmenta: Pipelined BFT Consensus with Slicing Broadcast -- 1 Introduction -- 2 Related Work -- 3 Analysis and Design -- 3.1 Directly Apply RBC Module in Hotstuff -- 3.2 Design to Reduce the Additional Communication Round -- 4 System Model -- 5 The Segmenta Protocol -- 5.1 Segmenta-Specific Data Structures -- 5.2 Segmenta Phases -- 5.3 Safety, Liveness, and Communication Complexity -- 5.4 Echoing Shards Optimistically -- 6 Chained Segmenta Protocol -- 7 Evaluation -- 7.1 Experimental Setup -- 7.2 Prediction Model -- 7.3 Throughput and Latency -- 8 Conclusion -- References -- Synthetic Data Generation for Differential Privacy Using Maximum Weight Matching -- 1 Introduction -- 2 Related Work -- 3 Preliminary -- 3.1 Problem Statement -- 3.2 Differential Privacy -- 4 Our Solution -- 4.1 Construction of Attribute Graph -- 4.2 Marginal Selection -- 4.3 PostProcess -- 4.4 Data Synthesis -- 5 Experiments -- 5.1 Experimental Settings -- 5.2 Experimental Results -- 6 Conclusion -- A Proof of Lemma -- References -- Approximate Multicast Coflow Scheduling in Reconfigurable Data Center Networks -- 1 Introduction.
2 Problem Formulation -- 2.1 System Model -- 2.2 Circuit Scheduling Model -- 2.3 Problem Formulation -- 3 The Delay-Efficient Multicast Coflow Scheduling Algorithm -- 3.1 Multicast Coflow Scheduling -- 3.2 Circuit Scheduling -- 3.3 Approximation Ratio Analysis -- 4 Numerical Evaluation -- 4.1 Simulation Settings -- 4.2 Performance Evaluation -- 5 Related Work -- 6 Conclusion -- References -- DAS: A DRAM-Based Annealing System for Solving Large-Scale Combinatorial Optimization Problems -- 1 Introduction -- 2 Background -- 2.1 Ising Model -- 2.2 The Overview of DRAM -- 2.3 In-DRAM Bulk Bitwise Operations -- 3 DAS: Detailed Design -- 3.1 An Overview of DAS -- 3.2 Spin Propagation Between Spin Nodes -- 3.3 Calculation of Local Search Term -- 3.4 Design of Probability Flipping -- 4 Evaluation and Applications -- 4.1 Area Overhead -- 4.2 Application and Analysis -- 5 Conclusion -- References -- Graph Neural Network for Critical Class Identification in Software System -- 1 Summary -- 2 Related Works -- 2.1 Research on the Importance of Nodes in Complex Networks -- 2.2 Research on Key Node Identification in Software Network -- 3 Basic Theory -- 3.1 Node Feature Extraction Based on Direct Extrapolation Graph Neural Network -- 3.2 Node Feature Extraction Based on Inductive Graph Neural Network -- 4 Research methods -- 4.1 Modeling Class dependency network -- 4.2 Network Embedding Learning -- 4.3 Key Class Identification -- 5 Experimental Analysis -- 5.1 Experiment Data -- 5.2 Experiment Setup -- 5.3 Analysis of Experimental Results -- 6 Conclusion -- References -- Spatio-Temporal Fusion Based Low-Loss Video Compression Algorithm for UAVs with Limited Processing Capability -- 1 Introduction -- 2 System Overview -- 3 Algorithm Design -- 3.1 Feature Clustering Based Temporal Video Sampling -- 3.2 Dynamic Encoding Based Spatial Video Sampling -- 4 Evaluation.
4.1 Experimental Setup -- 4.2 Experimental Results -- 5 Implementation -- 6 Related Work -- 7 Conclusion -- References -- CRAFT: Common Router Architecture for Throughput Optimization -- 1 Introduction -- 2 Related Work -- 3 The Proposed Viaduct Theory in Throughput Optimization -- 3.1 Viaduct Theory in Throughput Optimization -- 3.2 Comparison of Ring Decouple Theory and Viaduct Theory in Low-Cost Router -- 4 A Common Router Architecture for Throughput Optimization Based on Viaduct Theory -- 4.1 Flow Control Optimization with Viaduct Theory -- 4.2 Pipeline Stage and Latency Optimization -- 4.3 High Throughput Optimization with Viaduct Theory -- 4.4 Low Hardware Cost with Viaduct Theory -- 4.5 Bypass Strategy for Low Workload -- 5 Experiment Results -- 5.1 For Low-Cost Program -- 5.2 For High-Performance Program -- 5.3 Hardware Overhead -- 6 Conclusion -- References -- A Cross-Chain System Supports Verifiable Complete Data Provenance Queries -- 1 Introduction -- 2 Related Work -- 2.1 Query Function -- 2.2 Verifiable Query -- 2.3 Cross-Chain Technology -- 3 Preliminaries -- 3.1 Preliminary Knowledge -- 3.2 System Components -- 4 Solution -- 4.1 Verifiable Complete Data Provenance Query -- 4.2 Efficient Query -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Results and Discussion -- 6 Conclusion -- References -- Enabling Traffic-Differentiated Load Balancing for Datacenter Networks -- 1 Introduction -- 2 Motivation -- 3 TDLB Design -- 3.1 Design Overview -- 3.2 Design Rationale -- 3.3 Design Detail -- 4 Evaluation -- 4.1 Performce Under Symmetric Topology -- 4.2 Performance Under Asymmetric Topology -- 5 Related Work -- 6 Conclusion -- References -- Deep Reinforcement Learning Based Load Balancing for Heterogeneous Traffic in Datacenter Networks -- 1 Introduction -- 2 Background and Motivation -- 2.1 Background -- 2.2 Motivation -- 3 DRLB Design.
3.1 Design Overview -- 3.2 Design Detail -- 4 Experimental Results and Evaluation -- 4.1 Experimental Setup -- 4.2 Evaluation of Experimental Results -- 5 Related Works -- 6 Conclusion -- References -- Adaptive Routing for Datacenter Networks Using Ant Colony Optimization -- 1 Inroduction -- 2 Background and Motivation -- 2.1 Background -- 2.2 Motivation -- 2.3 Summary -- 3 RACO Design -- 3.1 RACO Overview -- 3.2 Statistics Module -- 3.3 Calculation Module -- 3.4 Algorithm Pseudocode -- 4 Performance Evaluation -- 4.1 Experimental Setup -- 4.2 Performance Under Symmetrical Topology -- 4.3 Performance Under Asymmetric Topology -- 5 Related Work -- 6 Conclusion -- References -- MPC: A Novel Internal Clustering Validity Index Based on Midpoint-Involved Distance -- 1 Introduction -- 2 MPdist: A Midpoint-Involved Distance -- 3 MPC: An Internal CVI Based on MPdist -- 3.1 The Definition of MPC -- 3.2 The Analysis of Time Complexity -- 4 Experimental Results -- 4.1 Description of the Test Datasets -- 4.2 Performances of CVIs on Evaluating the Results of the AHC -- 4.3 Performances of CVIs on Evaluating the Results of the K-Means -- 4.4 Performance Evaluation by Real Large-Scale Datasets -- 5 Conclusion and Future Works -- References -- HAECN: Hierarchical Automatic ECN Tuning with Ultra-Low Overhead in Datacenter Networks -- 1 Introduction -- 2 Motivation -- 3 HAECN Design -- 3.1 HAECN Overview -- 3.2 Design Details of HAECN -- 3.3 Reinforcement Learning Agent in HAECN -- 3.4 Learning Algorithm in HAECN -- 4 Implementation -- 5 Evaluation -- 5.1 Performance Under Bursty Networks -- 5.2 Performance Under Gradual Stabilization Network Environments -- 6 Related Work -- 7 Conclusion -- References -- An Android Malware Detection Method Based on Metapath Aggregated Graph Neural Network -- 1 Introduction -- 2 Related Work -- 3 Our Approach.
3.1 Feature Engineering.
Record Nr. UNISA-996587860503316
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. di Salerno
Opac: Controlla la disponibilità qui
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VI
Algorithms and Architectures for Parallel Processing : 23rd International Conference, ICA3PP 2023, Tianjin, China, October 20-22, 2023, Proceedings, Part VI
Autore Tari Zahir
Edizione [1st ed.]
Pubbl/distr/stampa Singapore : , : Springer, , 2024
Descrizione fisica 1 online resource (512 pages)
Altri autori (Persone) LiKeqiu
WuHongyi
Collana Lecture Notes in Computer Science Series
ISBN 981-9708-11-7
Formato Materiale a stampa
Livello bibliografico Monografia
Lingua di pubblicazione eng
Nota di contenuto Intro -- Preface -- Organization -- Contents - Part VI -- An Uncertainty-Aware Auction Mechanism for Federated Learning -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Definition -- 3.1 Federated Learning -- 3.2 Auction Framework -- 3.3 Problem Definition -- 4 The VCG-Based Auction Mechanism -- 4.1 Problem Transformation -- 4.2 Expected Social Welfare Maximization Algorithm -- 4.3 Payment -- 4.4 Properties -- 5 Greedy Algorithm Based Auction Mechanism -- 5.1 Choice of Clients' Training Data -- 5.2 Payment -- 5.3 Properties -- 5.4 Further Option for Reducing Time Complexity -- 6 Simulation Results -- 6.1 Experiment Settings -- 6.2 Experiment Analysis -- 7 Conclusion -- References -- Persistent Sketch: A Memory-Efficient and Robust Algorithm for Finding Top-k Persistent Flows -- 1 Introduction -- 2 Related Work -- 3 Problem Statement -- 4 The PE-Sketch Design -- 4.1 Approach Overview -- 4.2 Event Sampling Based on Flip Filter -- 4.3 Candidate Tracking -- 5 Performance Analysis -- 5.1 Parameters of Flip Filter -- 5.2 Estimation Error Bound of Candidate Matrix -- 6 Experimental Evaluation -- 6.1 Experiment Setup -- 6.2 Evaluation on Event Sampling -- 6.3 Finding Top-k Persistent Flows -- 6.4 Parameter Setting -- 7 Conclusion -- References -- FaCa: Fast Aware and Competition-Avoided Balancing for Data Center Network -- 1 Introduction -- 2 Related Work -- 2.1 Centralized Load Balancing -- 2.2 Distributed Load Balancing -- 3 Background -- 4 Motivation -- 4.1 Why Not ECMP? -- 4.2 A Motivation Example -- 5 Design -- 5.1 Overview -- 5.2 Network Model -- 5.3 Partly Global Vision with INT -- 5.4 Load Estimate -- 5.5 Flowing& -- Jumping -- 6 Implementation -- 6.1 Productive Experiment Setting -- 6.2 Implementation Challenges -- 7 Evaluation -- 7.1 Throughput Performance -- 7.2 Concurrent Competition -- 7.3 Micro-benchmark -- 8 Conclusion.
References -- Optimizing GNN Inference Processing on Very Long Vector Processor -- 1 Introduction -- 2 Background -- 2.1 GNN -- 2.2 Brief Introduction to the Long Vector Architecture Processor -- 3 Related Works -- 4 Efficient GNNs Inference on Long Vector Architecture -- 4.1 Efficient GNN Operator Library -- 4.2 GNN Operator Fusion Strategy -- 4.3 Heterogeneous Multi-threaded Cooperative Strategy for Long Vector Architecture -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Overall Performance -- 5.3 Operator Fusion Strategy -- 5.4 Heterogeneous Multi-threaded Cooperative Strategy -- 6 Conclusion -- References -- GDTM: Gaussian Differential Trust Mechanism for Optimal Recommender System -- 1 Introduction -- 2 Related Work -- 2.1 Trust-Based Recommendation -- 2.2 Recommended Protection Mechanism -- 3 Trade-Off Mechanism -- 3.1 Trust-Related Definition -- 3.2 Trust Recommendation -- 3.3 Gaussian Differential Trust Mechanism -- 4 Theoretical Analysis -- 5 Experiment Summary -- 5.1 Research Questions -- 5.2 Benchmark Model -- 5.3 Evaluation Metrics -- 5.4 Performance Analysis -- 6 Conclusion -- References -- Privacy-Enhanced Dynamic Symmetric Searchable Encryption with Efficient Searches Under Sparse Keywords -- 1 Introduction -- 2 Preliminaries -- 2.1 Notations -- 2.2 Dynamic Searchable Symmetric Encryption -- 3 Forward Privacy DSSE: SGsse-F -- 3.1 Construction of SGsse-F -- 3.2 An Example of SGsse-F -- 3.3 Correctness -- 3.4 Security Analysis -- 4 Forward and Backward Privacy DSSE: SGsse-FB -- 4.1 Correctness -- 4.2 Security Analysis -- 5 Related Works -- 6 Benefits and Drawbacks of DSSE Schemes Analysis -- 7 Evaluation -- 7.1 Experimental Setup -- 7.2 Experimental Analysis -- 8 Conclusion -- References -- CBA Sketch: A Sketching Algorithm Mining Persistent Batches in Data Streams -- 1 Introduction -- 1.1 Background and Motivation -- 1.2 Our Solution.
1.3 Key Contributions -- 2 Problem Formulation and Related Work -- 2.1 Problem Formulation -- 2.2 Related Work -- 3 The CBA Sketch -- 3.1 Circular-Time Sketch -- 3.2 Bloom Filter-Based Existence Recorder -- 3.3 Approximate Size Recorder -- 3.4 Query -- 4 Evaluation -- 4.1 Strawman Approach -- 4.2 Experimental Setup -- 4.3 Sensitivity Analysis -- 4.4 Experiments on ARE -- 4.5 Experiments on Throughput -- 5 Conclusion -- References -- Joint Optimization of System Bandwidth and Transmitting Power in Space-Air-Ground Integrated Mobile Edge Computing -- 1 Introduction -- 2 Related Works -- 2.1 Research on Air-Ground Communication -- 2.2 Research on Space-Ground Communication -- 2.3 Research on SAGIN Communication -- 3 System Model and Problem Formulation -- 3.1 Network Architecture -- 3.2 Satellite Coverage Time -- 3.3 IoRT Devices-Satellite/UAV Channel -- 3.4 Traffic Offloading Delay -- 3.5 Problem Formulation -- 4 Algorithm Design -- 4.1 DRL-Oriented Problem Transformation -- 4.2 SAC-Based Offloading Algorithm -- 5 Performance Evaluation -- 5.1 Parameter Setting -- 5.2 Performance Evaluation -- 6 Conclusion -- References -- A Novel Sensor Method for Dietary Detection -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Coupling Effect -- 3.2 Hardware Design -- 3.3 Event Detection Method Based on SVM -- 3.4 Chewing Counting Algorithm -- 4 Evaluation and Discussion -- 4.1 Experimental Methodology -- 4.2 The Evaluation of Chewing Event Detection -- 4.3 The Evaluation of Chewing Counting -- 5 Conclusion -- References -- Accelerated Optimization for Simulation of Brain Spiking Neural Network on GPGPUs -- 1 Introduction -- 2 Related Work -- 2.1 Spiking Neural Networks -- 2.2 SNN Simulation Tools -- 2.3 CUDA Programming on Multiple GPUs -- 3 Method -- 3.1 Workload-Oriented Partitioning -- 3.2 Memory Access Optimization.
3.3 Branch-Matched Thread Organization -- 3.4 Spike Synchronization Optimization -- 4 Experiments -- 4.1 Setup -- 4.2 Simulation Results -- 5 Conclusion -- References -- An Improved GPU Acceleration Framework for Smoothed Particle Hydrodynamics -- 1 Introduction -- 2 SPH Method -- 3 GPU Computation Using CUDA -- 3.1 Data Layout -- 3.2 CUDA Implementation -- 3.3 CUDA Acceleration -- 4 Results -- 4.1 GPU Simulation -- 4.2 Performance Evaluation -- 5 Conclusion and Future Work -- References -- MDCF: Multiple Dynamic Cuckoo Filters for LSM-Tree -- 1 Introduction -- 2 Background and Motivation -- 2.1 LSM-Tree Based KV Store -- 2.2 Serious Read Amplification of LSM-Tree -- 2.3 Motivation -- 3 Proposed Approach -- 3.1 The Design of MDCF -- 3.2 Maintenance of MDCF -- 3.3 The Flow of Read Operations -- 4 Evaluation -- 4.1 Experiment Setup -- 4.2 Workloads -- 4.3 Performance Evaluation -- 5 Related Work -- 6 Conclusion -- References -- A Game Theory Based Task Offloading Scheme for Maximizing Social Welfare in Edge Computing -- 1 Introduction -- 2 Motivation and Related Works -- 2.1 Motivation -- 2.2 Related Works -- 3 System Model -- 3.1 User Energy Consumption Model -- 3.2 User Utility Model -- 4 Stackelberg Game Based Offloading Algorithm -- 5 Evaluation -- 5.1 Evaluation Settings -- 5.2 Results -- 6 Conclusion -- References -- Research on Dos Attack Simulation and Detection in Low-Orbit Satellite Network -- 1 Introduction -- 1.1 A Subsection Sample -- 2 Related Work -- 3 Simulating DoS Attacks in Low Earth Orbit Satellite Networks -- 3.1 Satellite Physical Link Simulation Based on STK -- 3.2 Network Simulation Platform Based on OVS -- 3.3 SDN Control Platform Based on RYU -- 3.4 Network Attack Deployment Based on Docker Containers -- 4 Building and Detecting Datasets for Satellite Network Attacks -- 4.1 Overall Framework -- 4.2 Dataset Construction.
4.3 Stacking-Based Attack Detection -- 5 Experimental Results -- 6 Conclusion -- References -- Malware Detection Method Based on Visualization -- 1 Introduction -- 2 Related Work -- 2.1 Malware and Detection Methods -- 2.2 Visualization -- 2.3 Ensemble Machine Learning -- 3 Visualization-Based Malware Detection -- 3.1 Method Framework -- 3.2 Specific Implementation -- 4 Experiments and Evaluations -- 4.1 Data Set and Experimental Environment -- 4.2 Experimentation -- 5 Conclusion -- References -- TOC: Joint Task Offloading and Computation Reuse in Vehicular Edge Computing -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulation -- 3.1 System Model -- 3.2 Problem Formulation -- 4 TOC Framework -- 4.1 Computational Reuse -- 4.2 Minimum Weighted Vertex Search Algorithm -- 5 Experiment and Evaluation -- 5.1 Effects of Different Numbers of Tasks on Completion Time and Computational Execution Time on RSUs -- 5.2 Effects of Different Computation Capabilities on Completion Time and Computational Execution Time on RSUs -- 5.3 Lookup Rate Performance -- 6 Conclusion -- References -- Distributed Task Offloading for IoAV Using DDP-DQN -- 1 Introduction -- 2 Related Work -- 3 System Model -- 3.1 System Overview -- 3.2 Transmission Model -- 3.3 Task Delay -- 3.4 Energy Consumption Model -- 3.5 Formulation of Task Offloading Problem -- 4 DDP-DQN Algorithm in IoAV Offloading -- 4.1 IoAV Environment Model -- 4.2 DDP-DQN Based Task Offloading Algorithm -- 5 Performance Evaluation -- 5.1 Experimental Settings -- 5.2 Results and Analysis -- 6 Conclusion -- References -- A Task Offloading and Resource Allocation Optimization Method in End-Edge-Cloud Orchestrated Computing -- 1 Introduction -- 2 Related Work -- 3 System Model and Problem Formulation -- 3.1 Network Model -- 3.2 Communication Model -- 3.3 Computation Model -- 3.4 Problem Formulation.
4 DRL-Based Task Offloading and Resource Allocation in EECOC.
Record Nr. UNISA-996587861003316
Tari Zahir  
Singapore : , : Springer, , 2024
Materiale a stampa
Lo trovi qui: Univ. di Salerno
Opac: Controlla la disponibilità qui