
Neural Information Processing
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The three-volume set of LNCS 12532, 12533, and 12534 constitutes the proceedings of the 27th International Conference on Neural Information Processing, ICONIP 2020, held in Bangkok, Thailand, in November 2020. Due to COVID-19 pandemic the conference was held virtually.
The 187 full papers presented were carefully reviewed and selected from 618 submissions. The papers address the emerging topics of theoretical research, empirical studies, and applications of neural information processing techniques across different domains. The second volume, LNCS 12533, is organized in topical sections on computational intelligence; machine learning; robotics and control.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents - Part II
- Computational Intelligence
- A Novel Mathematic Entorhinal-Hippocampal System Building Cognitive Map
- 1 Introduction
- 2 Model Description
- 2.1 Architecture of the System
- 2.2 Neuron Model
- 2.3 Learning Algorithm
- 2.4 Visual Calibration
- 3 Experiment Result
- 3.1 Environment
- 3.2 Experiment for Cognitive Map Building
- 3.3 Experiment for Distribution of the Weight
- 3.4 Experiment for Property Verification
- 4 Discussion
- 4.1 Discussion on Cells and Parameters of the Model
- 4.2 Discussion on the Distribution of the Weight
- 4.3 Discussion on the Activities of Place Cells
- 5 Conclusion
- References
- Adaptive Risk-Return Control in Motor Planning
- 1 Introduction
- 2 Modeling
- 3 Model Evaluation
- 4 Summary and Discussion
- References
- Discrete Mother Tree Optimization for the Traveling Salesman Problem
- 1 Introduction
- 2 Discrete Mother Tree Algorithm (DMTO)
- 2.1 Solution Representation
- 2.2 Swap Operation
- 2.3 Updating the TMT
- 2.4 Updating the Solutions of the FPCTs Group
- 2.5 Updating the Solutions of FCTs Group
- 2.6 Updating the Solutions of LPCTs Group
- 2.7 DMTO Climate Change
- 3 Discrete Particle Swarm Optimization for TSPs
- 3.1 Discrete Particle Swarm Optimization (DPSO)
- 3.2 Updated Version of Discrete Particle Swarm Optimization (UDPSO)
- 4 Experimentation
- 5 Conclusion and Future Works
- References
- Dynamic Cloud Workflow Scheduling with a Heuristic-Based Encoding Genetic Algorithm
- 1 Introduction
- 2 Definition and Background
- 2.1 Workflow Modeling
- 2.2 Resource Modeling
- 2.3 Extended CPM
- 2.4 Objective Functions
- 3 Algorithm
- 3.1 Batch Processing
- 3.2 Scheduling Generator
- 3.3 Heuristic-Based Encoding GA
- 4 Experiments and Analysis
- 4.1 Efficiency of Single Heuristic Rules
- 4.2 Heuristic-Based GA vs. Pure GA
- 5 Conclusion
- References
- Multi-strategy Evolutionary Computation for Automated Jigsaw Puzzles
- 1 Introduction
- 2 Related Work
- 3 The Multi-strategy Evolution Algorithm
- 3.1 Compatibility Between Pieces and the Objective Function
- 3.2 Selector
- 3.3 Elite-Based Crossover
- 3.4 Four-Strategy Mutation
- 3.5 Diversity Enhancement Strategy
- 4 Experiments
- 4.1 Experimental Setup and Test Sets
- 4.2 Performance Metrics
- 4.3 Comparison Results
- 4.4 In-Depth Performance of MSEA
- 5 Conclusion
- References
- Real Valued Card Counting Strategies for the Game of Blackjack
- 1 Introduction
- 2 The Game of Blackjack
- 2.1 Card Counting Principles
- 2.2 Efficiency Metrics
- 2.3 Card Counting Strategies
- 3 Methods
- 3.1 Genetic Algorithm
- 3.2 Expert Advisor Mobile Application
- 4 Experiments and Results
- 4.1 Integer Weights
- 5 Conclusions
- References
- Machine Learning
- A Feature Selection Approach to Visual Domain Adaptation in Classification
- 1 Introduction
- 2 Related Work
- 3 A Feature Selection Approach to Visual Domain Adaptation in Classification
- 3.1 Particle Swarm Optimization
- 3.2 Fitness Function
- 3.3 Main Idea
- 4 Experiment Section
- 4.1 Data Preparation
- 4.2 State-of-the-Art Comparison Methods
- 4.3 Parameter Sensitivity and Experimental Setup
- 4.4 Comparative Analysis
- 5 Conclusions
- References
- A Framework for Reinforcement Learning with Autocorrelated Actions
- 1 Introduction
- 2 Related Work
- 2.1 Stochastic Dependence Between Actions
- 2.2 Reinforcement Learning with Experience Replay
- 3 Policy with Autocorrelated Actions
- 4 ACERAC: Actor-Critic with Experience Replay and Autocorrelated aCtions
- 4.1 Actor and Critic Training
- 5 Empirical Study
- 5.1 Experimental Setting
- 5.2 Results
- 5.3 Discussion
- 6 Conclusions and Future Work
- A Algorithms' Hyperparameters
- References
- A Motif-Based Graph Neural Network to Reciprocal Recommendation for Online Dating
- 1 Introduction
- 2 Related Work
- 2.1 Conventional RRS Approaches
- 2.2 Deep Learning Based RRS Approaches
- 2.3 Graph Neural Networks Based Approaches
- 3 The Proposed MotifGNN Approach
- 3.1 Preliminaries
- 3.2 Defined Motifs
- 3.3 Motif Based Random Walk Algorithm
- 3.4 Embedding with Attentive Graph Convolution
- 3.5 Reciprocal Recommendation Component
- 4 Experiments
- 4.1 Dataset and Experimental Settings
- 4.2 Model Performance Evaluation
- 4.3 Evaluation Results on Motif Effect
- 5 Conclusion
- References
- A Spiking Neural Architecture for Vector Quantization and Clustering
- 1 Introduction
- 2 Background
- 3 Methods
- 3.1 Neuron and Synapse Model
- 3.2 Input Encoding
- 3.3 Learning Rule
- 3.4 Lateral Interactions
- 4 Experiments and Results
- 4.1 Quality Assessment of the Reconstructed Images
- 4.2 Results on MNIST and Natural Images
- 5 Discussion
- References
- A Survey of Graph Curvature and Embedding in Non-Euclidean Spaces
- 1 Introduction
- 2 Mathematical Background
- 2.1 Preliminaries
- 2.2 Curvatures of Riemannian Geometry
- 2.3 Spaces of Constant Curvature
- 3 Graph Embeddings in Non-Euclidean Spaces
- 4 Applications and Tasks
- 5 Conclusion
- References
- A Tax Evasion Detection Method Based on Positive and Unlabeled Learning with Network Embedding Features
- 1 Introduction
- 2 Preliminary
- 3 PUNE Method
- 3.1 Problem Formulation
- 3.2 Extract Network Features
- 3.3 PU Training
- 4 Experiments
- 5 Conclusion
- References
- Adversarial Rectification Network for Scene Text Regularization
- 1 Introduction
- 2 Related Works
- 2.1 Generative Adversarial Networks
- 2.2 Scene Text Rectification and Recognition
- 3 Methodology
- 3.1 Network Architecture
- 3.2 Adversarial Learning
- 3.3 Loss Function for Rectifier
- 4 Experiments
- 4.1 Datasets
- 4.2 Implementation Details
- 4.3 Rectification Performance
- 4.4 Recognition Performance
- 5 Conclusion
- References
- An Overlapping Community Detection with Subspaces on Double-Views
- 1 Introduction
- 2 Related Work
- 3 Preliminaries
- 3.1 Problem Definition
- 3.2 The Construction of Double-Views
- 4 Methodology
- 4.1 The Objective Function of CDDV
- 4.2 Model Optimization
- 4.3 The Algorithm of CDDV
- 5 Experimental Evaluation
- 5.1 Experimental Settings
- 5.2 Experimental Results and Analysis
- 6 Conclusion and Future Work
- References
- API Based Discrimination of Ransomware and Benign Cryptographic Programs
- 1 Introduction
- 1.1 Command and Control Server Emulation
- 1.2 Contribution
- 2 Related Work
- 2.1 Ransomware Detection
- 3 Research Methodology
- 3.1 Cross-Validation Approach
- 3.2 C2 Emulators
- 4 Feature Selection and Classification
- 4.1 Feature Engineering
- 5 Experiments
- 5.1 Comparison with Existing Research
- 5.2 Experimental Setup
- 5.3 Feature Selection
- 5.4 Ransomware Against Benign Programs
- 5.5 Ransomware Against Benign-Cryptographic Programs
- 6 Conclusion
- References
- AutoGraph: Automated Graph Neural Network
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Problem Statement
- 3.2 Search Space
- 3.3 Evolutionary Algorithm
- 3.4 GNNs Evaluation
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Datasets
- 4.3 Baseline Methods
- 4.4 Results
- 5 Discussion and Conclusion
- References
- Automatic Curriculum Generation by Hierarchical Reinforcement Learning
- 1 Introduction
- 2 Related Works
- 3 Background
- 3.1 Multi-goal Reinforcement Learning
- 3.2 Deep Deterministic Policy Gradients
- 3.3 Hindsight Experience Replay
- 4 Method
- 4.1 Curriculum Generator
- 4.2 Action Policy
- 4.3 Algorithm Testing
- 5 Experiment
- 5.1 RL Environments
- 5.2 Experiment Details
- 5.3 Results and Comparison
- 5.4 Increasingly Difficult Curricula
- 5.5 Ablation Studies
- 6 Conclusion
- References
- Boltzmann Exploration for Deterministic Policy Optimization
- 1 Introduction
- 2 Related Work
- 3 Background
- 3.1 Markov Decision Process
- 3.2 Value Based Methods
- 3.3 Deep Deterministic Policy Gradient
- 4 Method
- 5 Experiments
- 5.1 Environments
- 5.2 Baseline Methods
- 5.3 Setup
- 5.4 Results and Analysis
- 6 Conclusion
- References
- Causal Inference for Mixed-Type Data in Additive Noise Models
- 1 Introduction
- 2 Model Definition
- 3 Model Estimation
- 3.1 Discrete Regression
- 3.2 Continuous Classification
- 3.3 Mixed Causal Inference
- 4 Related Work
- 5 Experiments
- 5.1 Evaluation on Synthetic Data
- 6 Conclusion
- References
- CDMC'19-The 10th International Cybersecurity Data Mining Competition
- 1 Introduction
- 2 The Activities of ICSDS
- 3 CDMC Annual Competition
- 3.1 Past Statistics
- 3.2 CDMC Cybersecurity Repository
- 3.3 Citation
- 4 CDMC 2019
- 4.1 Competition Process
- 4.2 Competition Tasks
- 4.3 Performance Evaluation
- 4.4 Results
- 5 Conclusion
- References
- Class-Balanced Loss for Scene Text Detection
- 1 Introduction
- 2 Methods
- 2.1 Class-Balanced Self Adaption Loss
- 2.2 Class-Balanced First Power Loss
- 3 Experiments
- 3.1 Datasets
- 3.2 Quantitative Results
- 3.3 Qualitative Results
- 4 Conclusion
- References
- Coordinated Behavior for Sequential Cooperative Task Using Two-Stage Reward Assignment with Decay
- 1 Introduction
- 2 Model and Problem
- 2.1 Models of Agents and Environment
- 2.2 Problem Formulation
- 2.3 Agents' Behaviors
- 3 Learning Methods
- 3.1 Deep Q-Network with Local Belief
- 3.2 Two-Stage Reward Assignment and Experience Replay
- 3.3 View Representation
- 3.4 Architecture of Neural Network
- 4 Experiments and Discussion
- 4.1 Experimental Setting
- 4.2 Performance Comparison
- 4.3 Analyzing Learned Coordinated Behavior
- 5 Conclusion
- References
- Deep Hierarchical Non-negative Matrix Factorization for Clustering Short Text
- 1 Introduction
- 2 Related Work
- 3 Deep Hierarchical NMF with SGNS-Based Embedding
- 3.1 Semantic Document Representation Learning with SGNS
- 3.2 Feature Learning with Hierarchical NMF
- 4 Empirical Analysis
- 5 Conclusion
- References
- Deep Reinforcement Learning with Temporal-Awareness Network
- 1 Introduction
- 2 Related Work
- 3 Deep Reinforcement Learning with Temporal-Awareness Network
- 3.1 Problem Formulation
- 3.2 DQN Without Temporal-Awareness
- 3.3 DQN with Temporal-Awareness
- 4 Experiments and Discussion
- 4.1 Experimental Setting
- 4.2 Comparison Experiments and Discussion
- 5 Conclusions
- References
- Double Replay Buffers with Restricted Gradient
- 1 Introduction
- 2 Related Work
- 3 Background
- 4 Method
- 4.1 Double Replay Buffers
- 4.2 Restricted Gradient
- 5 Experiment
- 5.1 Pendulum
- 5.2 MuJoCo
- 6 Conclusion
- References
- Entropy Repulsion for Semi-supervised Learning Against Class Mismatch
- 1 Introduction
- 2 Related Work
- 3 Our Method
- 3.1 Problem Formulation
- 3.2 Design Overview
- 3.3 Batch Annealing
- 3.4 Reloading with Temporal Pool
- 4 Evaluation
- 4.1 Experiment Configuration
- 4.2 Supervised with Mixup
- 4.3 ERCM-SSL Implementations
- 4.4 Results
- 4.5 Auxiliary Loss
- 4.6 Ablation Study
- 5 Conclusion
- References
- Estimating the Performance Indicators of Promotion Efficiency in FMCG Retail
- 1 Introduction
- 2 Problem Statement and Data Preparation
- 2.1 Feature Space of the Input Data
- 2.2 Data Preparation
- 3 Experiments
- 3.1 Prediction Model Pipeline
- 3.2 Performance Prediction of the KPI
- 3.3 Feature Importance
- 4 Conclusion and Discussion
- References
- Exploring User Trust and Reliability for Recommendation: A Hypergraph Ranking Approach
- 1 Introduction
- 2 Related Work
- 3 Preliminaries and Problem Statement
- 4 Weighting Strategy
- 4.1 Hyperedges Weighting
- 4.2 Vertices Weighting
- 5 Hypergraph Random Walk
- 6 Experiments
- 6.1 Datasets
- 6.2 Experimental Settings
- 6.3 Performance Comparison
- 6.4 Effect of Social Information
- 7 Conclusion
- References
- Facial Action Units Intensity Estimation via Graph Relation Network
- 1 Introduction
- 2 Related Work
- 2.1 Facial Action Units Analyses
- 2.2 Researchs on AU Relations
- 2.3 Graph Neural Networks
- 3 Proposed Methods
- 3.1 Graph Relation Layer
- 3.2 Multi-head Attention Strategy
- 3.3 Intensity Regression
- 3.4 Loss Function
- 4 Experiments
- 4.1 Settings
- 4.2 Results
- 5 Conclusion
- References
- Feature Selection Using Sparse Twin Bounded Support Vector Machine
- 1 Introduction
- 2 STBSVM
- 2.1 Notations
- 2.2 Objective Functions
- 2.3 Solutions and Property Analysis
- 3 Numerical Experiments
- 3.1 Toy Dataset
- 3.2 UCI Datasets
- 4 Conclusion
- References
- Few-Shot Classification with Transductive Data Clustering Transformation
- 1 Introduction
- 2 Methodology
- 2.1 Problem Statement
- 2.2 Review of Prototypical Network
- 2.3 Method
- 3 Experiment
- 3.1 Dataset
- 3.2 Implementation Detail
- 3.3 Results and Discussion
- 4 Conclusion
- References
- Forward Iterative Feature Selection Based on Laplacian Score
- 1 Introduction
- 2 Related Methods
- 2.1 Laplacian Score
- 2.2 Iterative Laplacian Score
- 3 Forward Iterative Feature Selection Based on Laplacian Score
- 4 Experimental Analysis
- 4.1 UCI Dataset
- 4.2 Microarray Gene Datasets
- 4.3 Statistical Comparison on Multiple Datasets
- 5 Conclusion
- References
- Functional Data Clustering Analysis via the Learning of Gaussian Processes with Wasserstein Distance
- 1 Introduction
- 2 Functional Data Clustering: Problem Formulation
- 3 Proposed Method
- 3.1 Gaussian Process Representation of Functional Data
- 3.2 Approximate Wasserstein Distance of Gaussian Processes
- 3.3 Barycenter Calculation
- 3.4 Clustering Algorithm and Extensions
- 4 Experimental Results
- 4.1 On Synthetic Data
- 4.2 On Real-World Data
- 5 Conclusions and Discussions
- References
- GPU-Based Self-Organizing Maps for Post-labeled Few-Shot Unsupervised Learning
- 1 Introduction
- 2 Proposed Methodology
- 2.1 Transfer Learning
- 2.2 Self-Organizing Maps Learning Procedure
- 2.3 SOM Labeling
- 3 Datasets and Implementation Details
- 3.1 mini-ImageNet Few-Shot Learning
- 3.2 WRN Training
- 3.3 Cosine Distance
- 4 SOM Software Implementation
- 4.1 TensorFlow-Based SOM
- 4.2 CPU and GPU Speedups
- 5 Experiments and Results
- 6 Discussion
- 7 Conclusion and Further Works
- References
- Gradient-Based Adversarial Image Forensics
- 1 Introduction
- 2 Background
- 3 The Proposed Method
- 3.1 Statistical Property of Adversarial Images
- 3.2 The Proposed Feature
- 4 Experimental Results
- 4.1 Experiment Settings
- 4.2 Compared with Previous Works
- 5 Conclusion
- References
- Hindsight-Combined and Hindsight-Prioritized Experience Replay
- 1 Introduction
- 2 Literature
- 2.1 Prioritized Experience Replay
- 2.2 Combined Experience Replay
- 2.3 Hindsight Experience Replay
- 3 Methodology
- 3.1 Replay Techniques
- 3.2 Learning Algorithm
- 3.3 Testing Environment
- 3.4 Machine Specifications
- 3.5 Investigation Outline
- 4 Results
- 4.1 The Hybrid Replay Algorithms
- 4.2 The Effect of Buffer Size Reduction
- 4.3 A Change in Environment
- 5 Conclusion
- References
- HPSGD: Hierarchical Parallel SGD with Stale Gradients Featuring
- 1 Introduction
- 2 Literature Review
- 3 Methodology
- 3.1 Implementation of Hierarchical Computation
- 3.2 Gradients Utilization
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Experiment Design and Analysis
- 5 Conclusions and Future Work
- References
- Implicit Posterior Sampling Reinforcement Learning for Continuous Control
- 1 Introduction
- 2 Background
- 2.1 Markov Decision Process
- 2.2 Deep Q Network
- 3 Methodology
- 3.1 Posterior Reinforcement Learning
- 3.2 Implicit Inference for Model Parameter
- 4 Experiments
- 4.1 Task Descriptions
- 4.2 Results and Analysis
- 5 Conclusion
- References
- Improving Multi-view Stereo with Contextual 2D-3D Skip Connection
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Main Net
- 3.2 2D-3D Skip Connection
- 3.3 Attention Combination Module
- 3.4 Network Training
- 4 Experiments
- 4.1 Dataset and Implementation Details
- 4.2 Results on ShapeNet Objects Reconstruction
- 4.3 Ablation Study
- 5 Conclusion
- References
- Improving Self-Organizing Maps with Unsupervised Feature Extraction
- 1 Introduction
- 2 Related Work and Methodology
- 2.1 Unsupervised Feature Extraction
- 2.2 Unsupervised Classification with Self-Organizing Maps (SOMs)
- 3 Implementation Details
- 3.1 CNN Training
- 3.2 SCAE Training
- 3.3 SNN Training
- 4 Experiments and Results
- 5 Discussion
- 6 Conclusion and Further Works
- References
- Information Security Implications of Machine-Learning-Based Automation in ITO Service Delivery - An Agency Theory Perspective*-6pt
- 1 Introduction
- 2 Related Work
- 2.1 ITO Service Transformation
- 2.2 ITO Service Delivery Automation Using ML
- 2.3 Lifecycle of MLA
- 3 Research Methodology
- 4 Findings and Analysis
- 4.1 Classification of Attack-Based ISR Factors in MLA
- 4.2 Adversarial Techniques on MLA in ITO
- 4.3 Implications of ISRs on ITO Service Delivery
- 5 Discussion and Future Directions
- 5.1 Current Constraints of ISRM in MLA of ITO Service Delivery
- 5.2 Theoretical Perspective
- 5.3 Limitations
- 6 Conclusion
- References
- Key Nodes Cluster Augmented Embedding for Heterogeneous Information Networks
- 1 Introduction
- 2 Related Work
- 3 Proposed Method
- 3.1 Problem Definition
- 3.2 Meta-path Based Embedding for HINs
- 3.3 Key Nodes Cluster Augmentation
- 3.4 A Unified Optimization Method
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Multi-label Classification
- 4.3 Link Prediction
- 4.4 Visualization
- 4.5 Parameter Analysis
- 5 Conclusion and Future Work
- References
- Processing of Incomplete Images by (Graph) Convolutional Neural Networks
- 1 Introduction
- 2 Graph-Based Model for Processing Incomplete Images
- 2.1 General Idea
- 2.2 Graph-Based Representation of Incomplete Images
- 2.3 Graph Convolutions
- 2.4 Spatial Graph Convolutions
- 2.5 Intuition Behind Spatial Graph Convolutions
- 3 Experiments
- 3.1 Reconstruction
- 3.2 Classification
- 4 Conclusion
- References
- Multi-agent Cooperation and Competition with Two-Level Attention Network
- 1 Introduction
- 2 Related Work
- 3 Background
- 3.1 Partially Observable Markov Game (POMG)
- 3.2 Proximal Policy Optimization (PPO)
- 4 Method
- 4.1 Across-Group Observation Attention Network (AGONet)
- 4.2 Intentional Communication Network (ICN)
- 4.3 Training
- 5 Simulations
- 5.1 Cooperation Navigation
- 5.2 Predator-Prey
- 6 Conclusion
- References
- Multi-view Subspace Adaptive Learning via Autoencoder and Attention
- 1 Introduction
- 2 Based Method
- 2.1 Self-representation of Data
- 2.2 MLRSSC
- 3 Our Proposed Framework
- 3.1 Encoder Layer
- 3.2 Multi-view Attention Layer
- 3.3 Self-representation Layer
- 3.4 Decoder Layer
- 3.5 Loss Function
- 4 Experimental Results
- 4.1 Datasets
- 4.2 Baseline Methods
- 4.3 The Configuration of Hyper-parameters
- 4.4 Experimental Results and Analysis
- 5 Conclusion and Future Work
- References
- Network Coding for Federated Learning Systems
- 1 Introduction
- 2 Related Work
- 2.1 Federated Learning (FL)
- 2.2 Network Coding (NC)
- 2.3 Differential Privacy (DP)
- 3 A Simple NC-FLS for Two Clients
- 3.1 Downlink Transmission
- 3.2 Uplink Transmission
- 4 Implementation
- 5 Experimental Results
- 5.1 Without Differential Privacy
- 5.2 With Differential Privacy
- 6 Conclusion and Future Work
- References
- New Approaches to Federated XGBoost Learning for Privacy-Preserving Data Analysis
- 1 Introduction
- 2 Preliminaries
- 2.1 XGBoost
- 2.2 Related Works
- 3 Overview of FL-XGBoost
- 3.1 Secure Model Updates Among Multiple Data Owners
- 3.2 FL-XGBoost with Random and Uniform Data Owner Selection
- 3.3 FL-XGBoost with Data Owner Selection Based on Prediction Confidence
- 4 Experiments
- 4.1 Data Set
- 4.2 Results
- 5 Security Analysis
- 5.1 Security Against the Central Server
- 5.2 Security Against the Data Owners
- 6 Conclusion
- References
- Partially Disentangled Latent Relations for Multi-label Deep Learning
- 1 Introduction
- 2 Problem Formulation
- 3 Partially Disentangled Latent Relations for Multi-label Deep Learning
- 3.1 The PDLRMDL for Multi-label Learning
- 3.2 Special Representation Learning
- 3.3 Invariant Special Representation to Translation and Rotation
- 3.4 The Diffusion Method for Transmitting the Instance Relations
- 3.5 The Self-Attention Layer for Multi-label Learning
- 3.6 The Proposed Objective Function
- 4 Experiments and Discussion
- 4.1 Datasets
- 4.2 Comparison Experiments and Discussion
- 5 Conclusion
- References
- Playing Catan with Cross-Dimensional Neural Network
- 1 Introduction
- 2 Background and Related Work
- 2.1 Deep Reinforcement Learning in Two-Player Games
- 2.2 Rules of Two-Player Catan
- 2.3 JSettlers and Research on Catan
- 3 Our Approach
- 3.1 Training Process
- 3.2 Encoding and Network Structure
- 4 Experiments and Results
- 4.1 Learning Curves
- 4.2 Long Term Training Results
- 4.3 Ablation Studies
- 5 Conclusion and Future Works
- References
- Port-Piece Embedding for Darknet Traffic Features and Clustering of Scan Attacks
- 1 Introduction
- 2 Port-Piece Embedding for Darknet Traffic Analysis
- 2.1 Creating Port-Sets
- 2.2 Port-Piece Embedding Vectors
- 2.3 Visualization of Scan Activities
- 2.4 Clustering of Scan Activities
- 3 Experiments
- 3.1 Similarity Measure of Port-Sets
- 3.2 Clustering of Scan Activities
- 3.3 Detection of New Vulnerabilities
- 4 Conclusions
- References
- Recency-Weighted Acceleration for Continuous Control Through Deep Reinforcement Learning
- 1 Introduction
- 2 Related Work
- 2.1 Value Estimation
- 2.2 Experience Replay
- 3 Preliminaries
- 3.1 Reinforcement Learning
- 3.2 DQN and Variants
- 3.3 Soft Actor-Critic
- 4 Recency-Weighted Acceleration Framework
- 4.1 Delayed Recency-Weighted Model
- 4.2 Phased-PER
- 5 Experiments
- 5.1 Experiment Setup
- 5.2 Comparative Evaluation
- 5.3 Ablation Studies
- 6 Conclusion
- References
- Regularized Multiset Neighborhood Correlation Analysis for Semi-paired Multiview Learning
- 1 Introduction
- 2 Background
- 2.1 Multiset Canonical Correlation Analysis
- 2.2 Semi-supervised Laplacian Regularization of MCCA
- 2.3 Semi-paired Learning of MCCA
- 3 Proposed Approach
- 3.1 Within-View Weight Matrix Construction
- 3.2 Cross-View Weight Matrix Construction
- 3.3 Model and Solution
- 4 Experiments
- 4.1 Parameter Selection
- 4.2 Results on the FERET Database
- 4.3 Results on the Yale Database
- 5 Conclusion
- References
- SAN: Sampling Adversarial Networks for Zero-Shot Learning
- 1 Related Work
- 2 Sampling Adversarial Networks
- 3 Experiments
- 3.1 Experimental Setup
- 3.2 Comparisons in Conventional Setting
- 3.3 Comparisons in Generalized Setting
- 3.4 Model Analysis
- 4 Conclusion
- References
- Semi-supervised Classification of Data Streams Based on Adaptive Density Peak Clustering
- 1 Introduction
- 2 Related Work
- 3 Proposed Algorithm
- 3.1 The Framework of the Proposed Algorithm
- 3.2 Adaptively Locate Cluster Centers and Label Unlabeled Instances
- 3.3 Concept Drift Detection
- 4 Experiments
- 4.1 Experimental Results
- 5 Conclusions
- References
- Stable Deep Reinforcement Learning Method by Predicting Uncertainty in Rewards as a Subtask
- 1 Introduction
- 2 Related Works
- 3 Method
- 3.1 Base Model: ABN-A3C
- 3.2 Variance Branch for Predicting Uncertainty in Rewards
- 4 Experiments
- 4.1 Model and Environment Setup
- 4.2 Evaluation Metrics
- 5 Results and Discussion
- 5.1 Atari Game Performance
- 5.2 Visualization of the Feature Map
- 5.3 Scalability
- 6 Conclusion
- References
- STGA-LSTM: A Spatial-Temporal Graph Attentional LSTM Scheme for Multi-agent Cooperation
- 1 Introduction
- 2 Related Works
- 2.1 Multi-agent Reinforcement Learning
- 2.2 Graph Convolution Network (GCN)
- 3 Preliminaries
- 3.1 Problem Definition
- 3.2 Partially Observable Markov Games
- 4 Method
- 4.1 Spatial Capture Network
- 4.2 Spatiotemporal LSTM Network
- 4.3 Policy Optimization
- 5 Simulations
- 5.1 Formation Control
- 5.2 Predator-Prey Games
- 6 Conclusions
- References
- SuperConv: Strengthening the Convolution Kernel via Weight Sharing
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 ACNet and Super-Kernel Convolution
- 3.2 How to Get the Weights of the Super-Kernel
- 3.3 SuperConv
- 3.4 Super-MixConv
- 4 Experiments
- 4.1 CIFAR-10
- 4.2 CIFAR-100
- 4.3 Ablation Studies
- 4.4 Discussions and Future Work
- 5 Conclusion
- References
- TAC-GAIL: A Multi-modal Imitation Learning Method
- 1 Introduction
- 2 Background
- 2.1 Preliminaries
- 2.2 Imitation Learning
- 2.3 Multi-modal Imitation Learning
- 3 Twin Auxiliary Classifiers GAIL
- 3.1 Insight of AC-GAIL
- 3.2 Twin Auxiliary Classifiers GAIL (TAC-GAIL)
- 4 Experiments
- 4.1 Experimental Environments
- 4.2 Experimental Setup
- 4.3 Results
- 5 Conclusion
- A Appendix
- A.1 Proof of Proposition2
- References
- Top-N Recommendation in P2P Lending: A Hybrid Graph Ranking Using Investor Profile
- 1 Introduction
- 2 Related Work
- 2.1 Graph-Based Approaches for Top-N Recommendation
- 2.2 Recommendation System in P2P Lending
- 3 Hybrid Graph Ranking Method
- 3.1 Product Profile
- 3.2 Investor Profile
- 3.3 Hybrid Graph Ranking Using Investor Profile
- 4 Experiments
- 4.1 Experiments Setting
- 4.2 Parameter Affect (RQ1)
- 4.3 Performance Comparison (RQ2)
- 4.4 Cold Start Affect (RQ3)
- 5 Conclusions and Future Work
- References
- WD3-MPER: A Method to Alleviate Approximation Bias in Actor-Critic
- 1 Introduction
- 2 Background
- 2.1 Reinforcement Learning
- 2.2 Double Q-Learning
- 2.3 DDPG
- 2.4 TD3
- 2.5 Per
- 3 Method
- 3.1 Weighted Double DDPG
- 3.2 MPER
- 3.3 WD3-MPER
- 4 Experiments
- 4.1 Experimental Evaluation
- 4.2 Experimental Comparison
- 5 Summary
- References
- Robotics and Control
- A Novel Vascular Robotic System: Performance Evaluation
- 1 Introduction
- 2 System Architecture
- 3 Experiments and Results
- 3.1 Experimental Setups
- 3.2 Experiment: Advancing the Guidewire at Different Speeds and Accelerations
- 3.3 Discussion
- 4 Conclusion
- References
- Accuracy Estimation for an Incrementally Learning Cooperative Inventory Assistant Robot
- 1 Introduction
- 2 Related Work
- 3 Accuracy Estimation with CGEM
- 4 Experimental Evaluation
- 4.1 Setup
- 4.2 Experiments
- 4.3 Results
- 5 Conclusion
- References
- Active Object Estimation for Human-Robot Collaborative Tasks
- 1 Introduction
- 2 Related Works
- 3 Methodology
- 3.1 Overview
- 4 Experiment
- 4.1 Dataset
- 4.2 Hyper-parameter Tuning
- 4.3 Experiment Settings
- 4.4 Quantitative Evaluations
- 4.5 Use Case
- 5 Conclusion
- References
- Adaptive Neural CPG-Based Control for a Soft Robotic Tentacle
- 1 Introduction
- 2 Soft Robotic Tentacle
- 3 Adaptive Neural Control and Implementation
- 3.1 Central Pattern Generator (CPG) for Periodic Movement Generation
- 3.2 Adaptation Mechanism for Online Movement Adaptation
- 4 Experimental Results
- 5 Discussion
- 6 Conclusion
- References
- Adaptive Neuromechanical Control for Robust Behaviors of Bio-Inspired Walking Robots
- 1 Introduction
- 2 Dung Beetle-Like Robot ALPHA
- 3 Adaptive Neuromechanical Control
- 3.1 Modular Neural Locomotion Control (MNC)
- 3.2 Online Adaptive Compliance Control (OAC)
- 4 Experiments and Results
- 5 Conclusions and Future Work
- References
- Deep Learning Based Strategy for Eye-to-Hand Robotic Tracking and Grabbing
- 1 Introduction
- 2 Framework of Moving Target Tracking and Grabbing Strategy
- 3 Target Recognition Based on YOLOv3 Algorithm
- 4 Target Tracking Algorithm
- 5 Experiments and Results
- 5.1 Train the YOLOv3 Network
- 5.2 Analysis of Recognition Results
- 5.3 Kalman Filter Target Trajectory Tracking Simulation
- 5.4 Experiment of Robotic Arm Tracking and Grabbing
- 6 Conclusion
- References
- Dynamical State Forcing on Central Pattern Generators for Efficient Robot Locomotion Control
- 1 Introduction
- 1.1 Related Works and Contributions
- 2 Central Pattern Generator
- 3 Dynamical State Forcing on CPG
- 4 Experiment
- 4.1 Investigation of Dynamical State Forcing CPG on a Single Motor
- 4.2 Investigation of Dynamical State Forcing on a Simulated Robot
- 5 Discussion
- References
- GSDCN: A Customized Two-Stage Neural Network for Benthonic Organism Detection
- 1 Introduction
- 2 Related Works
- 3 Methodology
- 3.1 Deformable Convolutional Module
- 3.2 Guided Anchoring Mechanism for Marine Organism
- 3.3 Sampling Balanced Strategy with Noise Labels
- 4 Experimental Results
- 4.1 Experimental Setting
- 4.2 Results
- 4.3 Ablation Experiments
- 5 Conclusion
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.