Neural Information Processing

Name: Neural Information Processing | 22nd International Conference, ICONIP 2015, Istanbul, Turkey, November 9-12, 2015, Proceedings, Part I
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

22nd International Conference, ICONIP 2015, Istanbul, Turkey, November 9-12, 2015, Proceedings, Part I

Sabri Arik Tingwen Huang Weng Kin Lai Qingshan Liu(Herausgeber*in)

Springer (Verlag)

Erschienen am 21. November 2015

XVII, 742 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-319-26532-2 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents - Part I
Texture Classification with Patch Autocorrelation Features
1 Introduction
2 Related Work
3 Translation and Rotation Invariant Patch Autocorrelation Features
3.1 Texture Features
4 Texture Classification Experiments
4.1 Data Set
4.2 Learning Methods
4.3 Implementation and Evaluation
4.4 Parameter Tuning
4.5 Results on Brodatz Data Set
5 Conclusion
References
Novel Architecture for Cellular Neural Network Suitable for High-Density Integration of Electron Dev ...
Abstract
1 Introduction
2 Device Architecture
2.1 Neuron
2.2 Synapse
2.3 Network
3 Learning Principle
4 Fabrication Process
5 Experimental Result
6 Conclusion
References
Analyzing the Impact of Feature Drifts in Streaming Learning
1 Introduction
2 Problem Statement
3 Simulating Feature Drifts
4 Analysis
4.1 Evaluated Algorithms
4.2 Experimental Protocol
4.3 Results Obtained
5 Conclusion
References
Non-linear Metric Learning Using Metric Tensor
Abstract
1 Introduction
2 Theoretical Analysis
3 Problem Simplification
4 Algorithm
5 Experiment
5.1 Performance in Supervised Metric Learning
5.2 Application in Semi-supervised Clustering
6 Conclusion
References
An Optimized Second Order Stochastic Learning Algorithm for Neural Network Training
1 Introduction
2 Proposed Algorithm
2.1 Overview of Learning Algorithms
2.2 Stochastic Diagonal Levenberg-Marquardt Algorithm
2.3 Bounded SDLM Algorithm
3 Experimental Design
4 Results and Discussions
5 Conclusion and Future Works
References
Max-Pooling Dropout for Regularization of Convolutional Neural Networks
Abstract
1 Introduction
2 Related Work
3 Max-Pooling Dropout
3.1 Max-Pooling Dropout at Training Time
3.2 Probabilistic Weighted Pooling at Test Time
4 Empirical Evaluations
4.1 Probabilistic Weighted Pooling vs. (Scaled) Max-Pooling
4.2 Max-Pooling Dropout vs. Stochastic Pooling
5 Conclusions
References
Predicting Box Office Receipts of Movies with Pruned Random Forest
1 Introduction
2 Methodology
2.1 Movie Information Data Collection
2.2 Pruned Random Forest
2.3 Advice for Screen Schedule
3 Results
3.1 The Classification Performance of Pruned Random Forest
3.2 Comparison with Other Models
4 Conclusion
References
A Novel 1-graph Based Image Classification Algorithm
1 Introduction
2 Background
2.1 Sparse Representation Based Classification Algorithm
2.2 1-Graph
3 1-graph Based Image Classification Method
3.1 Relationship Between Training Samples and Classes
3.2 Classification Process
4 Experiment Results
4.1 Face Recognition
4.2 Handwritten Digit Recognition
5 Conclusion and Future Work
References
Classification of Keystroke Patterns for User Identification in a Pressure-Based Typing Biometrics S ...
Abstract
1 Introduction
2 System Design
2.1 Force Sensor
2.2 Microprocessor Design with Arduino
3 Classification
3.1 Particle Swarm Optimization
3.2 K-Means
4 Experimental Setup and Results
5 Conclusions
References
Discriminative Orthonormal Dictionary Learning for Fast Low-Rank Representation
1 Introduction
2 Discriminative Orthonormal Dictionary Learning
2.1 Formulation
2.2 Optimization
3 Fast Low-Rank Representation
4 Experiments
4.1 Extended Yale B Dataset
4.2 AR Dataset
4.3 Caltech 101 Dataset
5 Conclusions
References
Supervised Topic Classification for Modeling a Hierarchical Conference Structure
1 Introduction
2 Supervised Classification, Flat Case
3 Topics Hierarchy
4 Empirical Results
5 Conclusion
References
A Framework for Online Inter-subjects Classification in Endogenous Brain-Computer Interfaces
Abstract
1 Introduction
2 Methods
2.1 Base Classifiers' Weights Initialization
2.2 Base Classifiers' Weights Adaptation Using Ensemble Predictions
2.3 Base Classifiers' Weights Adaptation Using Ensemble Predictions Reinforced by Interaction Error-Related Potentials
3 Experiments
3.1 EEG Data Sets
3.2 Procedure for Simulating IErrPs
3.3 Results
4 Conclusion
References
A Bayesian Sarsa Learning Algorithm with Bandit-Based Method
1 Introduction
2 Bayesian Sarsa
2.1 Q-values Distribution
2.2 Updating Q-Values
2.3 Actions Selection
3 Experimental Results
3.1 Gridworld
4 Conclusion
References
Incrementally Built Dictionary Learning for Sparse Representation
1 Introduction
2 Background on Dictionary Learning
3 Incrementally Built Dictionary Learning
3.1 Approach Description
3.2 Incremental Learning Rule
3.3 Sparse Coding-Based Feature Extraction
4 Experimentations
4.1 Digits Recognition
4.2 Face Recognition
4.3 The Effects of Incremental Learning
5 Conclusion
References
Learning to Reconstruct 3D Structure from Object Motion
1 Introduction
2 Related Work
3 DNN Based 3D Reconstruction Method
3.1 Reconstruction Unit
3.2 Deep Neural Network for 3D Reconstruction
3.3 Temporal Integration
4 Experiments
4.1 Data Generation
4.2 Reconstruction on Synthetic Images
4.3 Reconstruction on Real Images
5 Conclusions
References
Convolutional Networks Based Edge Detector Learned via Contrast Sensitivity Function
1 Introduction
2 The Model Architecture
2.1 Convolutional Networks
2.2 Multi-channel Structure
3 Training Data Generation and Annotation
3.1 Training Data Generation
3.2 Training Data Annotation
4 Experiments
5 Conclusion
References
Learning Algorithms and Frame Signatures for Video Similarity Ranking
1 Introduction
2 Similar-Video Retrieval
2.1 Frame Features
2.2 Clustering Algorithms for Exemplar Extraction
2.3 Global and Local Alignments
3 Video Signature Tools
3.1 Frame Signature
3.2 Word and Bag of Words
4 Experiments on Video Similarity Ranking
4.1 Test Video Set and Evaluation Method
4.2 Experimental Results
5 Discussions
References
On Measuring the Complexity of Classification Problems
1 Introduction
2 Complexity Measures/Indices
2.1 Feature/Attribute Overlapping
2.2 Separability of Classes
2.3 Geometry, Topology and Density
3 Conclusion
References
The Effect of Stemming and Stop-Word-Removal on Automatic Text Classification in Turkish Language
Abstract
1 Introduction
2 Related Work
3 Proposed Work
4 Methodology for Dataset
5 The Experimental Results
6 Conclusion
References
Example-Specific Density Based Matching Kernel for Classification of Varying Length Patterns of Speech Using Support Vector Machines
1 Introduction
2 Dynamic Kernels for Sets of Feature Vectors
3 Example-Specific Density Based Matching Kernel for Sets of Feature Vectors
4 Experimental Studies on Speech Emotion Recognition and Speaker Identification
5 Conclusions
References
Possibilistic Information Retrieval Model Based on Relevant Annotations and Expanded Classification
Abstract
1 Introduction
2 Related Works
3 Filtering Annotation Approach
4 Classification of Annotations
4.1 Initial Classification
4.2 Clusters Extension
5 Experimental Evaluation and Analysis of Results
5.1 Used Collection of Data
5.2 Effects of the Classified and Filtered Annotation
6 Conclusion and Future Works
References
A Transfer Learning Method with Deep Convolutional Neural Network for Diffuse Lung Disease Classification
1 Introduction
2 Methods
2.1 Deep Convolutional Neural Network (DCNN)
2.2 Transfer Learning for DCNN
2.3 Materials
3 Results
4 Summary and Discussion
References
Evaluation of Machine Learning Algorithms for Automatic Modulation Recognition
Abstract
1 Introduction
2 System Model, Signal and Channel Representation
3 Feature Extraction
3.1 Spectral Features
3.2 Statistical Features
4 Nonnegative Matrix Factorization (NMF)
5 Experimental Results
6 Conclusion
References
Probabilistic Prediction in Multiclass Classification Derived for Flexible Text-Prompted Speaker Verification
1 Introduction
2 Probabilistic Prediction for Text-Prompted Speaker Verification
2.1 Multistep Speaker and Text Verification Using GEBI
2.2 Probabilistic Prediction for Speaker and Text Verification
2.3 Loss Functions for Evaluating the Performance
3 Experiments
3.1 Experimental Setting
3.2 Experimental Results and Analysis
4 Conclusion
References
Simple Feature Quantities for Learning of Dynamic Binary Neural Networks
1 Introduction
2 Dynamic Binary Neural Networks
3 Teacher Signal and Feature Quantities
4 Greedy Search Based Sparsification Algorithm
5 Conclusions
References
Transfer Metric Learning for Kinship Verification with Locality-Constrained Sparse Features
1 Introduction
2 Proposed Approach
2.1 Feature Extraction
2.2 NRTML
3 Experiments
3.1 Experimental Settings
3.2 Experimental Results
4 Conclusion
References
Unsupervised Land Classification by Self-organizing Map Utilizing the Ensemble Variance Information in Satellite-Borne Polarimetric Synthetic Aperture Radar
1 Introduction
2 Stokes Vector and Poincare Sphere Parameter
3 Unsupervised Learning Classification of Land States by Using SOM Based on Ensemble Variance
3.1 Two-Stage Clustering to Utilize the Dispersion Feature
3.2 Step 1: Local Clustering by Paying Attention to the Cluster Variance
3.3 Step 2: Classification of Clusters in SOM
4 Experimental Results
5 Conclusion
References
Algorithmic Robustness for Semi-Supervised (, , )-Good Metric Learning
1 Introduction
2 Notations and Related Work
3 Learning Consistent Good Similarity Functions
3.1 Optimization Problem
3.2 Consistency Guarantees
4 Experiments
4.1 Experimental Setup
4.2 Results
5 Conclusion
References
Patchwise Tracking via Spatio-Temporal Constraint-Based Sparse Representation and Multiple-Instance Learning-Based SVM
1 Introduction
2 Patchwise Tracking via a Hybrid Generative-Discriminative Appearance Model
2.1 Generative Appearance Model Based on STSR
2.2 Discriminative Appearance Model Based on MIL&SVM
2.3 Adaptive Hybrid Generative-Discriminative Appearance Model
3 Experiments
4 Conclusion
References
An Autonomous Mobile Robot with Functions of Action Learning, Memorizing, Recall and Identifying the ...
Abstract
1 Introduction
2 Whole Structure of the System
3 Action Learning
4 State-Action Storage System
4.1 Chaotic Neural Network (CNN)
4.2 Memorizing and Recall of an Environmental Data Using Mutual Associative Type CNN
5 Identification of the Environment
5.1 Identification Method of the Environment
5.1.1 Mean and Variance Configuration Algorithm
5.1.2 Identification Algorithm of the Environment
6 Computer Simulation
6.1 Preparation
6.2 Observation State Description
6.3 Simulation 1
6.4 Simulation 2
7 Conclusion
References
SEIR Immune Strategy for Instance Weighted Naive Bayes Classification
1 Introduction
2 SEIR Immune Strategy Based Instance Weighted Bayes
2.1 Instance Weighted Naive Bayes
2.2 SEIR Immune Strategy for Instance Weighting
2.3 SWNB Classifier
3 Experiments
3.1 Experimental Conditions and Baselines
3.2 UCI Standard Classification Task
3.3 Convergence and Learning Curves
4 Conclusion and Future Work
References
Enhancing Competitive Island Cooperative Neuro-Evolution Through Backpropagation for Pattern Classification
1 Introduction
2 Proposed Method
2.1 Backpropagation in CICN
2.2 Backpropagation Island
3 Experiments and Results
3.1 Results and Discussion
4 Conclusions and Future Work
References
Email Personalization and User Profiling Using RANSAC Multi Model Response Regression Based Optimize ...
Abstract
1 Introduction
2 Email Personalization
2.1 Gradient Boost Trees
2.2 Optimized Extreme Learning Machines as Base Estimators for Gradient Boosting Trees
2.2.1 RANSAC Multi Model Response Regularization
3 Experimental Results
4 Conclusions
References
An Auto-Encoder for Learning Conversation Representation Using LSTM
Abstract
1 Introduction
2 The Dataset for Our Work
3 The Auto-Encoder Model
3.1 LSTM
3.2 The Interactive Scheme in LSTM-Encoder
3.3 The Computation in LSTM-Decoder
3.4 Training
4 Experiments
4.1 Experimental Setting
4.2 Experimental Results and Analysis
5 Conclusions and Future Work
References
On the Use of Score Ratio with Distance-Based Classifiers in Biometric Signature Recognition
1 Introduction
2 Score Ratio
3 Experimental Setup
3.1 Score Normalization
3.2 Corpus
3.3 Experimental Sets
3.4 Signature Verification Systems
3.5 Performance Measure
4 Results
4.1 Vector Quantization-Based System (VQSys)
4.2 Dynamic Time Warping-Based System (DTWSys)
4.3 Fractional Distance-Based System (FraDisSys)
4.4 Results Analysis
5 Conclusions and Future Works
References
A Multifactor Dimensionality Reduction Based Associative Classification for Detecting SNP Interactions
Abstract
1 Introduction
2 Methods
2.1 MDR Based Associative Classifier (MDRAC)
2.2 Data Simulation
2.3 Sporadic Breast Cancer Data
3 Results
3.1 Analysis of Simulated Data
3.2 Analysis of Breast Cancer Data
4 Conclusions and Future Work
References
Distributed Q-learning Controller for a Multi-Intersection Traffic Network
1 Introduction
2 Q-learning for Traffic Signal Timing
2.1 The Q-learning Algorithm
2.2 Proposed Developed Q-learning Controller
3 Experiments Environment and Results Discussion
4 Conclusions
References
Learning Rule for Linear Multilayer Feedforward ANN by Boosted Decision Stumps
1 Introduction
2 Generalized Boostron
2.1 Learning Weights of an Output Neuron
2.2 Learning Weights of a Hidden Neuron
2.3 Learning a Multiclass Classifier
3 Experiments and Results
3.1 Datasets Description and Experimental Settings
3.2 Results
4 Conclusions
References
Class-Semantic Color-Texture Textons for Vegetation Classification
Abstract
1 Introduction
2 Related Work
3 Proposed Approach
3.1 Color and Texture Feature Extraction
3.2 Class-Semantic Color-Texture Texton Generation
3.3 Texton Occurrence and Superpixel-Based Voting
4 Experiments and Comparative Analysis
4.1 Evaluation Datasets
4.2 Global Accuracy
4.3 Global Accuracy Versus Size of Gaussian Filters
4.4 Class Accuracy
4.5 Performance Comparisons
5 Conclusions
Acknowledgments
References
Towards Unsupervised Learning for Arabic Handwritten Recognition Using Deep Architectures
Abstract
1 Introduction
2 Deep Belief Networks (DBN)
3 Convolutional Neural Networks (CNN)
4 Experimental Results and Discussion
4.1 HACDB Database
4.2 Experiments Using Deep Belief Network
4.3 Experiments Using Convolutional Neural Networks
5 Discussion
6 Conclusion
References
Optimum Colour Space Selection for Ulcerated Regions Using Statistical Analysis and Classification o ...
Abstract
1 Introduction
1.1 Wireless Capsule Endoscopy
1.2 Problem Definition: Ulcer
2 Related Works
2.1 Colour Space Transformation and Information Content
2.2 Machine Learning
3 Methodology
3.1 Image Enhancement
3.2 Colour Space Selection for Enhanced Representation of Ulcer Information
4 Experimental Results and Discussions
4.1 Data Set
4.2 Results of Statistical Analysis
5 Conclusion
References
Learning the Optimal Product Design Through History
1 Introduction
2 The Proposed Method
2.1 Computing Novelty
2.2 Computing Performance
3 Computational Experiments
4 Conclusion
References
Learning Shape-Driven Segmentation Based on Neural Network and Sparse Reconstruction Toward Automated Cell Analysis of Cervical Smears
1 Introduction
2 Methodology
2.1 Cellular Components Classification Based on Feed Forward Neural Network
2.2 Individual Cytoplasm Segmentation
3 Material and Experimental Setting
4 Experimental Results
5 Conclusions
References
Adaptive Differential Evolution Based Feature Selection and Parameter Optimization for Advised SVM C ...
Abstract
1 Introduction
2 Proposed Model
3 Details of Proposed Algorithms
3.1 Adaptive Differential Evolution
3.2 Advised Support Vector Machine Based Classification
4 Experimental Results
5 Conclusion and Future Work
References
TNorm: An Unsupervised Batch Effects Correction Method for Gene Expression Data Classification
Abstract
1 Introduction
2 The Proposed Method
3 Experimental Design
3.1 Gene Expression Datasets
3.2 Gene Expression Analysis
3.3 Batch Effects Correction
3.4 Cross Dataset Classification
4 Results and Discussion
5 Conclusion
References
Finger-Vein Quality Assessment by Representation Learning from Binary Images
Abstract
1 Introduction
2 DNN for Finger-Vein Image Quality Assessment
2.1 Deep Neural Networks
2.2 Feature Extraction
2.3 Generating Quality Score
3 Experiments
3.1 Database
3.2 Experiment Settings on Database A
3.3 Experiment Settings on Database B
3.4 Selection of CNN Architecture
3.5 Evaluation of Quality Image Assessment
3.6 Effects on the Finger-Vein Verification System
4 Conclusions and Future Work
Acknowledgments
References
Learning to Predict Where People Look with Tensor-Based Multi-view Learning
1 Introduction
2 Algebraic Framework
3 The Optimisation Problem
4 Performance Evaluations
4.1 Randomly Select Missing Views
4.2 Fixing a Missing View
5 Conclusion
References
Classification of the Scripts in Medieval Documents from Balkan Region by Run-Length Texture Analysis
1 Introduction
2 The Algorithm
2.1 The Proposed Algorithm
2.2 Mapping According to Typographical Features
2.3 Feature Extraction with the Run-Length Statistics
2.4 Feature Extraction by GA-ICDA
3 Experimentation, Results and Discussion
4 Conclusion
References
Accelerating Artificial Bee Colony Algorithm for Global Optimization
1 Introduction
2 Basic ABC Algorithm
3 Our Approach
3.1 Trigonometric Search Equation in the Onlooker Bee Phase
3.2 Orthogonal Learning in the Scout Bee Phase
3.3 The Procedure of Our Approach
4 Experiments
4.1 Benchmark Functions
4.2 Comparison with Other ABC Variants
5 Conclusions
References
Classification of High and Low Intelligent Individuals Using Pupil and Eye Blink
Abstract
1 Introduction
2 Bio-Signal Analysis for Determining Intelligence
3 Proposed Method
3.1 Experimental Setup and Data Acquisition
3.2 Feature Extraction
3.3 Classification of High and Low Intelligent Individuals
4 Experimental Results
4.1 Participants
4.2 Statistical Analysis
4.3 Comparison with Other Classifiers
5 Conclusion and Future Work
References
Learning Task Specific Distributed Paragraph Representations Using a 2-Tier Convolutional Neural Network
1 Introduction
2 Our Approach
2.1 Distributed Word Representation Model
2.2 Distributed Sentence Representation Tier
2.3 Distributed Paragraph Representation Tier
3 Evaluation and Discussion
3.1 Experiment Settings
3.2 DBpedia Ontology Classification
3.3 Amazon Review Sentiment Analysis
3.4 Discussion
4 Conclusion and Future Directions
References
A Comparison of Supervised Learning Techniques for Clustering
1 Introduction to Data Mining and Classification
2 A Brief Overview of the Classification Techniques
2.1 Naive Bayes
2.2 Support Vector Machine
2.3 Decision Tree
2.4 Random Forest
3 The Analysis
3.1 Computational Algorithm
3.2 Data Analysis
3.3 Results
4 Summary and Conclusions
References
Radar Pattern Classification Based on Class Probability Output Networks
1 Introduction
2 Key Features for Radar Pattern Classification
3 Class Probability Output Networks for Emitter Identification
4 Simulation
5 Conclusion
References
Hierarchical Data Classification Using Deep Neural Networks
1 Introduction
2 History of Deep Neural Networks
3 Data Representation
4 Experimental Results and Discussion
5 Conclusion and Future Work
References
Model Inclusive Learning for Shape from Shading with Simultaneously Estimating Illumination Directions
1 Introduction
2 Model Inclusive Learning for Shape from Shading-Simultaneous Estimation of Illumination Directions
3 Problem Formulation and Proposed Learning Method
3.1 Problem Formulation
3.2 Proposed Learning Method
4 Experiments
4.1 Experiment 1 - Synthetic Image
4.2 Experiment 2 - Real Image
5 Conclusion
References
A Computational Model of Match Decision-Making Problem Using Spiking SHESN with Reward-Modulated Reinforcement Learning
Abstract
1 Introduction
2 Problem Description
3 Methods
3.1 Network Architecture
3.2 Reward-Modulated Reinforcement Learning for Output Synapses
4 Results and Discussion
4.1 Construction of Training Datasets and Test Datasets
4.2 The Behavioral Results of Model
5 Discussion and Conclusion
Acknowledgment
References
Identify Website Personality by Using Unsupervised Learning Based on Quantitative Website Elements
Abstract
1 Introduction
2 Mapping WPS Items to Quantitative Website Elements
3 Method
4 Data Extraction
5 Experiment Results and Analysis
6 Summary and Future Work
References
Discriminative Dictionary Learning for Skeletal Action Recognition
1 Introduction
2 Representation of Trajectory Information
3 Projective Dictionary Pair Learning (PDPL)
4 Experimental Results
4.1 CAD-60 Dataset
4.2 MSR Daily Activity3D Dataset
5 Conclusion
References
Single Face Image Super-Resolution via Multi-dictionary Bayesian Non-parametric Learning
Abstract
1 Introduction
2 Problem Formulation
3 Proposed Approach
3.1 Training Set Clustering
3.2 Dictionary Learning with Beta Process
3.3 Synthesizing High-Resolution Face Image
4 Experimental and Discussion
4.1 Training Set Clustering
4.2 Non-parametric Dictionary Learning
4.3 Super Resolution Result
5 Conclusions
Acknowledgements
References
Sparse LS-SVM in the Sorted Empirical Feature Space for Pattern Classification
1 Introduction
2 Sparse Least Squares Support Vector Training in the Reduced Empirical Feature Space
2.1 Reduced Empirical Feature Space
2.2 Training SLS-SVM in REF
3 Sparse Least Squares Support Vector Training in the Sorted Empirical Feature Space
4 Experimental Results
4.1 Parameter Setting
4.2 Performance Comparison
5 Conclusions
References
A Cost Sensitive Minimal Learning Machine for Pattern Classification
1 Introduction
2 Minimal Learning Machine
2.1 Distance Regression
2.2 Output Estimation
3 Weighted Minimal Learning Machine
3.1 Imbalanced Data Classification
3.2 Classification with Reject Option
4 Results
4.1 Imbalanced Classification
4.2 Classification with Reject Option
5 Conclusion
References
A Minimal Learning Machine for Datasets with Missing Values
1 Introduction
2 Minimal Learning Machine
2.1 Distance Regression
2.2 Output Estimation
3 Computing Dx in the Presence of Missing Data
3.1 Expected Squared Distance (ESD) Calculation
4 Performance Evaluation
5 Results
6 Conclusion
References
Calibrated k-labelsets for Ensemble Multi-label Classification
1 Introduction
2 Related Work
3 Calibrated k-labelsets Ensemble Method
4 Experimental Evaluation
4.1 Data Sets and Experimental Setup
4.2 Results
5 Conclusion
References
EMG Signal Based Knee Joint Angle Estimation of Flexion and Extension with Extreme Learning Machine ...
Abstract
1 Introduction
2 Proposed Model
3 Data Acquisition
4 Signal Processing
5 Proposed Knee Model
6 Data Analysis
7 Conclusion
References
Continuous User Authentication Using Machine Learning on Touch Dynamics
1 Introduction
2 Previous Work
3 Proposed Approach
3.1 Dataset
3.2 Feature Processing
3.3 Models and Tools
4 Results
5 Conclusions
6 Future Work
References
Information Theoretical Analysis of Deep Learning Representations
1 Introduction
2 Materials and Methods
2.1 MNIST Database
2.2 Deep Neural Network
2.3 Learning Algorithms
2.4 Evaluation
3 Result
4 Discussion
4.1 Entropy, Mutual Information and Performance
4.2 Differences in Pretraining Algorithms
5 Conclusion
References
Hybedrized NSGA-II and MOEA/D with Harmony Search Algorithm to Solve Multi-objective Optimization Problems
1 Introduction
2 Methodology
2.1 The Proposed Hybridized Frameworks
3 Experimental Results
4 Conclusion
References
A Complex Network-Based Anytime Data Stream Clustering Algorithm
1 Introduction
2 Related Work
2.1 CluStream
2.2 ClusTree
2.3 DenStream
3 Complex Networks
4 CNDenStream
5 Experimental Evaluation
5.1 Discussion
5.2 Parameter Sensitivity
6 Conclusion
References
Robust Online Multi-object Tracking by Maximum a Posteriori Estimation with Sequential Trajectory Prior
1 Introduction
2 Our Approach
2.1 Problem Formulation
2.2 Detection Refinement with MAP Estimation
2.3 Data Association with MAP Estimation
3 Experiments
3.1 Implementation Details
3.2 Datasets and Object Detector
3.3 Evaluation Metrics
3.4 Results and Discussion
4 Conclusion
References
Enhance Differential Evolution Algorithm Based on Novel Mutation Strategy and Parameter Control Method
1 Introduction
2 Classical Differential Evolution Algorithm
2.1 Initialization
2.2 Mutation Operator
2.3 Crossover Operator
2.4 Selection Operator
3 The Enhanced Differential Evolution Algorithm
3.1 The New Mutation Strategy
3.2 The New Parameter Control Method
3.3 The Enhance Differential Evolution Algorithm (EDE)
4 Experiments and Results
5 Conclusion
References
Hybrid Model for the Training of Interval Type-2 Fuzzy Logic System
1 Introduction
2 Structure of the Interval Type-2 FLS Used in This Paper
3 Structure of the Hybrid Learning Model for Interval Type-2 FLS (IT2FELM-GA)
3.1 Optimal Parameters Using GA
3.2 Extreme Learning Machine Strategy for Interval Type-2 FLS (IT2FELM) [24]
3.3 Objective Function Evaluation
3.4 GA Operations
4 Simulation Results
5 Conclusion
References
A Numerical Optimization Algorithm Based on Bacterial Reproduction
1 Introduction
2 Design of BRO Algorithm
2.1 The Frame of BRO
2.2 BRO Algorithm Design
3 Experimental Results and Analysis
3.1 Comparative Algorithms Parameters Setting
3.2 Experimental Results and Analysis
4 The Theoretical Analysis of BRO
4.1 The Analysis of Convergence
4.2 The Analysis of Time Complexity
5 Conclusion
References
Visual-Textual Late Semantic Fusion Using Deep Neural Network for Document Categorization
1 Introduction
2 DNN Late Semantic Fusion
3 Implementation
4 Experimental Evaluation
5 Conclusions
References
Prototype Selection on Large and Streaming Data
1 Introduction
2 Methodology
2.1 MCNN Rule
2.2 Prototype Selection on Remaining Data Points
2.3 Algorithm
3 Results
4 Conclusion
References
GOS-IL: A Generalized Over-Sampling Based Online Imbalanced Learning Framework
1 Introduction
2 Proposed Framework: GOS-IL
3 Experimental Study
3.1 Data Sets
3.2 Methods and Learner Settings
3.3 Experimental Results
4 Conclusion
References
A New Version of the Dendritic Cell Immune Algorithm Based on the K-Nearest Neighbors
1 Introduction
2 Problem Statement
3 The Proposed Approach: A New Dendritic Cell Algorithm Based on the K-Nearest Neighbors
3.1 The KNN-DCA Architecture
3.2 The KNN-DCA Classification Phase
4 Experimental Setup and Results
5 Conclusion and Future Directions
References
Impact of Base Partitions on Multi-objective and Traditional Ensemble Clustering Algorithms
1 Introduction
2 Related Work
3 Materials and Methods
4 Experimental Protocol
4.1 Performance Evaluation
4.2 Parameters' Settings
4.3 Sets of Base Partitions
5 Results
6 Final Remarks
References
Multi-Manifold Matrix Tri-Factorization for Text Data Clustering
1 Introduction
2 Related Works and Aims
3 Multi-Manifold Matrix Tri-Factorization Based Co-clustering
3.1 Problem Formalization
3.2 Multi-Manifold Co-clustering Algorithm (MMC)
3.3 Optimization
4 Numerical Experiments
4.1 Parameter Settings
4.2 Results
5 Conclusion
References
Clustering of Binary Data Sets Using Artificial Ants Algorithm
1 Introduction
2 Related Work
3 Swarm Intelligence and Clustering
4 The CL-Ant Clustering Algorithm
5 Experimental evaluation
5.1 Experiments with Binary Data Sets
6 Conclusion and Future Works
References
Inverse Reinforcement Learning Based on Behaviors of a Learning Agent
1 Introduction
2 Background
2.1 Markov Decision Process (MDP)
2.2 Reinforcement Learning
2.3 Related Works
3 Inverse Reinforcement Learning with a Developing Agent
3.1 Likelihood of Reward Function
3.2 Matrix Representation of SARSA
4 Numerical Experiment
5 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken