Neural Information Processing

Name: Neural Information Processing | 25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part IV
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part IV

Long Cheng Andrew Chi Sing Leung Seiichi Ozawa(Editor)

Springer (Publisher)

Published on 3. December 2018

XXII, 699 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-04212-7 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
ICONIP 2018 Organization
Contents - Part IV
Feature Selection
Multi-label Feature Selection Method Combining Unbiased Hilbert-Schmidt Independence Criterion with Controlled Genetic Algorithm
1 Introduction
2 Novel Multi-label Feature Selection Method
2.1 Multi-label Classification and Feature Selection Settings
2.2 Hilbert-Schmidt Independence Criterion
2.3 Controlled Genetic Algorithm
2.4 Multi-label Feature Selection Method Based on HSIC and CGA
3 Experiments
3.1 Four Benchmark Data Sets
3.2 Compared Methods and Experimental Settings
3.3 Convergence Analysis for CGAHSIC
3.4 Experimental Results and Analysis
4 Conclusions
References
Anthropometric Features Based Gait Pattern Prediction Using Random Forest for Patient-Specific Gait Training
1 Introduction
2 Method
2.1 A Lower Limb Rehabilitation Robot
2.2 Anthropometric Features and Gait Trajectory Fitting
2.3 Feature Selection Based on mRMR
2.4 The Random Forest Model
3 The Results and Discussion
3.1 The Result of Fourier Series Fitting
3.2 Feature Selection and Optimization of the RF Model
3.3 Predicted Performance of the RF Models
4 Conclusion
References
Robust Multi-view Features Fusion Method Based on CNMF
1 Introduction
2 Robust Multi-view Subspace Fusion Method
2.1 CMF
2.2 RCNMF
2.3 Solution to RCNMF
2.4 Algorithm and Its Complexity Analysis
3 Experimental Results and Analysis
3.1 Compared Methods
3.2 Effect of Parameters
3.3 Cluster Results Analysis
4 Conclusion
References
Brain Functional Connectivity Analysis and Crucial Channel Selection Using Channel-Wise CNN
1 Introduction
2 Methods
2.1 Short Time Fourier Transformation
2.2 Design of the CWCNN Model
3 Experiments and Results
3.1 Emotional Dataset
3.2 Data Processing and Model Training
3.3 Experiment Results
4 Conclusion
References
An Effective Discriminative Learning Approach for Emotion-Specific Features Using Deep Neural Networks
1 Introduction
2 Overall Architecture
2.1 Frame-Level Acoustic Feature Extraction
2.2 Segment-Level Emotion Classification
2.3 Utterance-Level Feature and Classification
3 Loss
3.1 Softmax Loss
3.2 Combining with Centre Loss
3.3 Optimisation
4 Experiments and Results
4.1 General Experimental Setting
4.2 Experiment on CASIA Chinese Emotional Corpus
4.3 Experiment on Berlin Emo-DB Corpus
4.4 Experiment on SAVEE Database
4.5 Experiment on Hyper-parameters and
5 Conclusions
References
Convolutional Neural Network with Spectrogram and Perceptual Features for Speech Emotion Recognition
1 Introduction
2 Baseline System
3 CNN Based on Spectrogram and Perceptual Features
3.1 Motivation for Fusing Spectrogram and Perceptual Features
3.2 Fusion Strategy of CSF and RSF
4 Experiment
4.1 Experimental Setup
4.2 Experimental Results
5 Conclusions and Future Works
References
Feature Selection Based on Fuzzy Conditional Distinction Degree
1 Introduction
2 Preliminaries
3 Fuzzy Conditional Distinction Degree
4 Experiments
4.1 Datasets and Experimental Setup
4.2 Monotonicity Experiment
4.3 Redundancy Experiment
4.4 Classification Experiment
5 Conclusion
References
Multi-label Feature Selection Method Based on Multivariate Mutual Information and Particle Swarm Optimization
1 Introduction
2 Multi-label Feature Selection Algorithm Based on Multivariate Mutual Information and Particle Swarm Optimization
2.1 Max-relevance and Min-redundancy Criterion Based on Multivariate Mutual Information
2.2 Binary Particle Swarm Optimization with Two Mutation Operations
2.3 Multi-label Feature Selection Method Based on MMI and M2BPSO
3 Experiments
3.1 Four Benchmark Data Sets
3.2 Compared Methods and Experimental Settings
3.3 Performance Evaluation and Analysis
4 Conclusions
References
Feature Selection Using Distance from Classification Boundary and Monte Carlo Simulation
Abstract
1 Introduction
2 Method
2.1 Pseudosample Exploration in Kernel Space
2.2 Preliminary Experiment for Pseudosample Exploration
2.3 Core Procedure for Feature Selection
3 Results
3.1 Dataset
3.2 Performance Comparison of Algorithms
4 Conclusion
References
Clustering
Approximate Spectral Clustering Using Topology Preserving Methods and Local Scaling
Abstract
1 Introduction
2 Related Work
2.1 Similarity Measures for Spectral Clustering
2.2 Approximate Spectral Clustering
3 Proposed Approach
3.1 Approximating Input Data
3.2 Constructing the Affinity Matrix {\varvec A}
3.3 Embedding Space Dimensions {\mathbb{R}}^{m \times k}
3.4 Number of Clusters in the Embedding Space
4 Experiments
4.1 Synthetic Data
4.2 UCI Datasets
4.3 Berkeley Segmentation Dataset (BSDS500)
5 Conclusions
References
Discovering Similarities in Malware Behaviors by Clustering of API Call Sequences
1 Introduction
2 Related Work
3 Proposed Malware Clustering Framework
3.1 Generating Encoded API Call Sequences
3.2 Generating Distance Matrix
3.3 Clustering
4 Results and Discussions
4.1 Clusters of Similar Malware Behavioral Patterns
4.2 Clustering Quality
5 Conclusion
References
A Storm-Based Parallel Clustering Algorithm of Streaming Data
Abstract
1 Introduction
2 Clustering Algorithm
2.1 Data Preprocessing
2.2 Single-Pass Clustering Algorithm Principle
3 Storm-Based Parallel Clustering Algorithm
3.1 Storm Framework
3.2 Parallel Clustering Algorithm
4 Emulation Results and Analysis
4.1 Emulation Environment and Data Set
4.2 Emulation Results and Analysis
5 Conclusions
References
Iterative Maximum Clique Clustering Based Detection Filter
1 Introduction
2 Related Work
3 Iterative Clustering Based Detection Filter
3.1 Iterative Maximum Clique Clustering
3.2 Detection Filter
3.3 Ground-Perception Filtering Method
4 Experiments
4.1 Analysis of Iterative Clustering
4.2 Comparison with Baseline
4.3 Application in Tracking
5 Conclusion
References
Towards a Compact and Effective Representation for Datasets with Inhomogeneous Clusters
Abstract
1 Introduction
2 Review of Boundary Extraction
3 Robust Boundary Extraction
4 Experiment
5 Conclusion
References
Adaptive Fuzzy Clustering Algorithm with Local Information and Markov Random Field for Image Segmentation
1 Introduction
2 Method
2.1 HMRF-FCM Algorithm
2.2 Proposed Method
3 Experiment
3.1 Comparison Experiments on Synthetic Images
3.2 Comparison Experiments on Natural Image
3.3 Comparison Experiments on Brain MR Images
4 Conclusion
References
Efficient Direct Structured Subspace Clustering
1 Introduction
2 Related Work
3 Methodology
3.1 Efficient Direct Structured Subspace Clustering
3.2 Optimization
4 Experiment
4.1 Evaluation Metrics
4.2 Experimental Analysis
5 Conclusion
References
Privacy-Preserving K-Means Clustering Upon Negative Databases
1 Introduction
2 Related Work
2.1 Privacy Preserving K-Means Clustering
2.2 Negative Database
2.3 The K-Hidden Algorithm
3 Euclidean Distance Estimation for Negative Databases
3.1 Theoretical Analyses
3.2 Experiments on Euclidean Distance Estimation Error
4 Privacy-Preserving K-Means Algorithm upon NDBs
5 Experiments
5.1 Experiments on the Impact of r
5.2 Experiments on the Impact of p
6 Conclusions
References
Self-Paced Multi-Task Multi-View Capped-norm Clustering
1 Introduction
2 Preliminaries
2.1 Multi-Task Multi-View Clustering
2.2 Self-Paced Learning
3 Self-Paced Multi-Task Multi-View Capped-norm Clustering
3.1 The Objective Function
3.2 Optimization
3.3 Time Complexity Analysis
4 Experimental Setup
4.1 Data Sets
4.2 Baseline Methods
4.3 Evaluation Measure
4.4 Parameter Setting
5 Results and Analysis
5.1 Clustering Results on Real Data
5.2 Sensitivity Analysis
6 Conclusion
References
Shape Clustering as a Type of Procrustes Analysis
1 Introduction
2 Preliminaries
2.1 Configuration Matrix
2.2 Ordinary Procrustes Sum of Squares
2.3 Full Procrustes Mean
3 Main Results
3.1 Competitive Learning Scheme
3.2 Aim of the Clustering Algorithm
4 Computational Experiments
4.1 Datasets
4.2 Assessment
4.3 Results
5 Conclusion
References
Classification
Aspect-Level Sentiment Classification with Conv-Attention Mechanism
1 Introduction
2 Related Work
3 Model
3.1 The Multilayer Attention Architecture
3.2 Conv-Attention Mechanism
3.3 Model Training
4 Experiments
4.1 Experiment Setting
4.2 Impacts of Attention Layer and Filter Size
4.3 Model Comparisons
4.4 Computing Time
4.5 The Effect of Conv-Attention
4.6 Case Study
5 Conclusion
References
Attention-Based Combination of CNN and RNN for Relation Classification
1 Introduction
2 Related Work
3 Proposed Model
3.1 Structure Overview
3.2 Data Noise Removing Based on SDP
3.3 SDP-Based Attention
3.4 Attention-Based Pooling
3.5 Training Objective
4 Experiments
4.1 Dataset
4.2 Settings
4.3 Results
4.4 Analysis
5 Conclusion
References
Discrete Sparse Hashing for Cross-Modal Similarity Search
1 Introduction
2 Method
2.1 Discrete Sparse Hashing
2.2 Optimization Algorithm
2.3 Learning Hash Functions
3 Experiments
3.1 Datasets Experimental Settings
3.2 Results and Discussions
3.3 Convergence Analysis
3.4 Parameter Sensitivity Analysis
4 Conclusion
References
Solving the Double Dummy Bridge Problem with Shallow Autoencoders
1 Introduction
2 Autoencoder-Based Architectures
2.1 A Deal Representation in the Input Layer
2.2 Hidden (Feature) Layers
2.3 Output Layer
2.4 Topology of Connections
2.5 Network Architectures
3 Experimental Setup and Results
3.1 Results
4 Summary and Future Work
References
Cross-Project Issue Classification Based on Ensemble Modeling in a Social Coding World
1 Introduction
2 Background and Related Work
2.1 Issue Reports Classification
2.2 Cross-Project Prediction
3 Approach
4 Dataset
4.1 Data Collection
4.2 Filtering
4.3 Preprocessing
5 Experiment
5.1 Model Training
5.2 Experimental Methodology
5.3 Experimental Design
5.4 Evaluation Metrics
6 Results
7 Conclusion
References
Decision Tree Twin Support Vector Machine Based on Kernel Clustering for Multi-class Classification
Abstract
1 Introduction
2 Twin Support Vector Machine
3 Decision Tree Twin Support Vector Machine Based on Kernel Clustering
3.1 Construction of Decision Tree
3.2 Algorithm Description
4 Experimental Results
4.1 Vehicle Dataset
4.2 Experimental Results
5 Conclusions
Acknowledgements
References
Machine Learning Techniques for Classification of Livestock Behavior
Abstract
1 Introduction
2 Methodology
2.1 Dataset
2.2 Data Preparation
2.3 Feature Extraction
2.4 Exploratory Analysis
2.5 Feature Selection
2.6 Classification Algorithms
2.7 Evaluation Metrics
2.8 Model Training Using 10-Fold Cross Validation
3 Results
4 Discussion and Conclusion
5 Future Work
Acknowledgements
References
Two-Stage Attention Network for Aspect-Level Sentiment Classification
1 Introduction
2 Related Works
3 Backgrounds
3.1 Long Short-Term Memory (LSTM)
3.2 Attention Mechanism
4 Two-Stage Attention-Based LSTM
4.1 Structure of Our Model
4.2 Preprocess Layer
4.3 Two-Stage Attention Layer
5 Experiment
5.1 Comparison to Other Methods
5.2 Case Study
6 Conclusions and Future Works
References
The Fuzzy Misclassification Analysis with Deep Neural Network for Handling Class Noise Problem
Abstract
1 Introduction
2 Related Work
3 Proposed Method
3.1 Fuzzy Misclassification Analysis
3.2 Assigning Fuzzy Membership Values
3.3 Deep Neural Network (DNN)
4 Experimental Design
4.1 The Characteristic of the Datasets
4.2 Experimental Processes
4.3 Experimental Results
5 Conclusion
Acknowledgement
References
A Neuronal Morphology Classification Approach Based on Deep Residual Neural Networks
Abstract
1 Introduction
2 Human Neuron Data Sets
3 Neuronal Morphology Classification Approach
3.1 Deep Residual Neural Networks
3.2 Classification Approach of 3D Neurons
4 Experimental Results and Analysis
4.1 Experimental Parameter Settings
4.2 Neuron Classification Process Analysis
4.3 Classification Performance Comparison
5 Conclusion
Acknowledgment
References
Privacy-Preserving Naive Bayes Classification Using Fully Homomorphic Encryption
1 Introduction
2 Related Works
2.1 Preparation
2.2 Li's PP-NBC Model
3 Proposed Privacy-Preserving Naive Bayes Classifier
3.1 Data Representation
3.2 Protocol 1
3.3 Protocol 2
3.4 Protocol 3
4 Experiments
4.1 Data Set
4.2 Prediction Accuracy
4.3 Execution Time
5 Conclusions
References
Classification of Calligraphy Style Based on Convolutional Neural Network
1 Introduction
2 Related Work
3 Methods
3.1 Local Convolution Neural Network
3.2 Global Convolution Neural Network
3.3 Two Pathway Convolution Neural Network
4 Experiments
4.1 Classification of Calligraphy Styles
4.2 Comparison to Related Work
5 Conclusion
References
Tropical Fruits Classification Using an AlexNet-Type Convolutional Neural Network and Image Augmentation
1 Introduction
2 Literature Review
3 Methodology
3.1 Fruit-AlexNet CNN
3.2 Image Augmentation
4 Results and Discussion
5 Conclusion
References
Supervised and Semi-supervised Multi-task Binary Classification
1 Introduction
2 Supervised Multi-task Binary Gaussian Process Classification
3 Semi-supervised Multi-task Binary Gaussian Process Classification
4 Experiments
4.1 English Alphabets Binary Classification
4.2 USPS Digit Binary Classification
4.3 Evaluation Measures
5 Conclusion
References
Employ Decision Values for Soft-Classifier Evaluation with Crispy References
1 Introduction
2 Preliminary
3 Evaluating Soft Classifiers with Crispy References
3.1 The Multi-class Case
3.2 The Binary Case
4 Experiment
4.1 On Artificial Chessboard Data
4.2 On Real-World Data
5 Conclusion
References
Detection
Guide-Wire Detecting Based on Speeded up Robust Features for Percutaneous Coronary Intervention
1 Introduction
2 Method
2.1 Preprocessing
2.2 SURF
2.3 Detection Method
3 Results
4 Conclusion
References
New Default Box Strategy of SSD for Small Target Detection
1 Introduction
2 A New Default Box Strategy of SSD for Small Target Detection
2.1 New Scales Based on the Change in Size of the Six Layers
2.2 New Aspect Ratios with Fewer Default Boxes
3 Experiment and Analysis
3.1 Small Ground Target Dataset
3.2 Performance Comparison of the Proposed Strategy with the Original Default Box Strategy
3.3 Influence of the Scale Range on the Detection Precision of SSD
3.4 Influence of Different Aspect Ratios on the Performance of SSD
4 Conclusion
References
Weakly Supervised Temporal Action Detection with Shot-Based Temporal Pooling Network
1 Introduction
2 Our Approach
2.1 Two-Stream Network for Feature Extraction
2.2 Shot-Based Sampling Method
2.3 Temporal Pooling Network
2.4 Training Procedure
2.5 Detection Generation and Prediction
3 Experiments
3.1 Dataset and Setup
3.2 Evaluation on Visual Feature Encoder
3.3 Evaluation on Attention Module
3.4 Evaluation on Shot-Based Sampling Method
3.5 Comparison with the State-of-the-Art Methods
4 Conclusion
References
Drogue Detection for Autonomous Aerial Refueling Based on Adaboost and Convolutional Neural Networks
Abstract
1 Introduction
2 Structure and Basic Model
2.1 Framework Overall
2.2 Adaboost Algorithm
2.3 Tiny CNN Model
3 Improved Focal Loss
4 Experimental Result
4.1 Dataset
4.2 Detailed Settings
4.3 Results
5 Conclusions
References
Deep Neural Network Based Salient Object Detection with Image Enhancement
Abstract
1 Introduction
2 Proposed Model
2.1 Network Architecture
2.2 Image Enhancement Network
2.3 Saliency Prediction Network
3 Experimental Results
3.1 Experiment Settings
3.2 State-of-the-Art Performance Comparison
3.3 Contribution of Image Enhancement Network
4 Conclusion
Acknowledgments
References
Brain Slices Microscopic Detection Using Simplified SSD with Cycle-GAN Data Augmentation
1 Introduction
2 Related Works
2.1 Data Augmentation
2.2 Microscopic Object Detection
3 Data Augmentation
3.1 Data Augmentation with Traditional Techniques
3.2 Data Augmentation with Cycle-GAN
4 Simplified SSD Detection Model
4.1 Network Architecture
4.2 Training
5 Experiments
5.1 Data
5.2 Results
6 Conclusions
References
Agglomeration Detection in Gas-Phase Ethylene Polymerization Based on Multi-scale Convolutional Neural Network
Abstract
1 Introduction
2 MCNN Methodology
2.1 MCNN Architecture
2.2 Classification Based on MCNN
3 Related Experimental Work
3.1 Experimental Device and Experimental Data
3.2 Implementing the MCNN Computational Graph Using TensorFlow
4 Experimental Results
4.1 Experimental Data and Experimental Sample Selection
4.2 Agglomeration Detection Based on MCNN
5 Conclusion
Acknowledgements
References
Intra-class Structure Aware Networks for Screen Defect Detection
1 Introduction
2 Intra-class Structure Aware Networks
2.1 Intra-class Variant Embedded Space
2.2 The Regularization for Cluster Assignments of Intra-class Variant
2.3 Optimization
3 Experiments
3.1 Experimental Setup
3.2 Experimental Results
4 Conclusion
References
Mobile Malware Detection - An Analysis of the Impact of Feature Categories
1 Introduction
2 Characteristics of Feature Categories
3 Encoding Contextual Information as Feature Value
4 Experiments and Results
4.1 Dataset
4.2 Performance Measure
4.3 Impact of Feature Categories and Finding Best Combination of Feature Categories
4.4 Verification Using Different Classifiers
4.5 Further Analysis of ICC Features
5 Conclusion and Future Work
References
Recurrent RetinaNet: A Video Object Detection Model Based on Focal Loss
1 Introduction
2 Background
3 Recurrent RetinaNet
3.1 Backbone
3.2 Feature Pyramid Network
3.3 Anchors
3.4 Regression Subnet
3.5 Classification Subnet
3.6 Detection
4 Experiments
4.1 Experiment Setup
4.2 Performance and Analysis
5 Conclusion
References
Neural Causality Detection for Multi-dimensional Point Processes
1 Introduction
2 A Review of Multi-dimensional Point Process
2.1 Multi-dimensional Point Process
3 The Proposed Approach
3.1 Recurrent Multi-dimensional Temporal Point Process
3.2 Discrete RMDTPP
3.3 Granger Causality
4 Experiment
4.1 Dataset and Evaluation Metric
4.2 Experiment Results
5 Conclusion
References
Density-Induced Support Vector Data Description for Fault Detection on Tennessee Eastman Process
1 Introduction
2 Density-Induced Support Vector Data Description
2.1 Methods for Extracting Relative Density Degrees
2.2 Density-Induced SVDD
3 D-SVDD for Fault Detection
3.1 Analysis on Parameter T
3.2 D-SVDD for Fault Detection
4 Experiments
4.1 Experiments on Parameter T
4.2 Comparison with Other Methods
5 Conclusions
References
ExtTra: Short-Term Traffic Flow Prediction Based on Extremely Randomized Trees
1 Introduction
2 Related Work
3 Proposed Approach
3.1 Data Set
3.2 General Framework
3.3 Feature Extraction
3.4 Extremely Randomized Trees
4 Experimental Evaluation
4.1 Experiment Setup
4.2 Experimental Results
5 Conclusion
References
Actor Model Anomaly Detection Using Kernel Principal Component Analysis
Abstract
1 Introduction
2 Introduction
2.1 Feature Extraction
2.2 Data Clustering
2.3 Actor Model Anomaly Detection Algorithm
3 The Experimental Results
3.1 Experimental Data Collections
3.2 Experimental Results Display
4 Conclusions
References
Passive Detection of Splicing and Copy-Move Attacks in Image Forgery
Abstract
1 Introduction
2 Related Works
3 Proposed Method
3.1 Converting Images into Gray Scale
3.2 Block Division of Input Image
3.3 Block Discrete Cosine Transformation (BDCT)
3.4 Local Binary Pattern (LBP) Operator
3.5 Block Division of LBP Image
3.6 Apply SRIV and Feature Generation
4 Experiments and Results
4.1 Description of Datasets
4.2 SVM Classifier and Model Validation
4.3 Results and Discussion
4.4 Comparison with Recent Methods
5 Conclusion
Acknowledgement
References
Learning Latent Byte-Level Feature Representation for Malware Detection
1 Introduction and Related Work
2 Sparse Hashing Algorithm
3 Binary Word2vec Feature Representation
4 Experiments and Results
4.1 Results and Discussion
5 Conclusion
References
Occlusion Detection in Visual Tracking: A New Framework and A New Benchmark
1 Introduction
2 Occlusion Detection Framework
2.1 COD Review
2.2 Adaptive COD
3 Occlusion Benchmark
4 Experiments
4.1 Quantitative Evaluation
4.2 Qualitative Evaluation
5 Conclusion
References
Attentional Payload Anomaly Detector for Web Applications
1 Introduction
2 Related Works
2.1 Feature Extraction and Feature Selection
2.2 Deep Learning
2.3 Attention Mechanism
3 ATPAD Model
3.1 Hidden State Generation
3.2 Attention Calculation
3.3 Payload Classification
3.4 Attention Visualization
4 Experiment
4.1 Dataset
4.2 Configurations of ATPAD
4.3 How Does the Attention Help?
4.4 Performance Evaluation on CSIC-2010
5 Conclusion
References
A Semantic Parsing Based LSTM Model for Intrusion Detection
1 Introduction
2 Related Work
3 Proposed Model
3.1 Data Preprogress
3.2 LSTM Sequence Labeling Model for Intrusion Detection
4 Experiment and Results
4.1 Performance Metric
4.2 Comparative Analysis
5 Conclusion
References
Detecting the Doubt Effect and Subjective Beliefs Using Neural Networks and Observers' Pupillary Responses
Abstract
1 Introduction
2 Hypotheses
3 Experimental Design
3.1 Stimuli
3.2 Participants
3.3 Measure and Sensor
3.4 Procedure
4 Methodology
4.1 Feature Extraction
4.2 Model Description
5 Results
5.1 Veracity Judgments
5.2 Pupillary Responses
5.3 Relationship Between Pupillary Responses and Veracity Judgments
5.4 Relationship Between Pupillary Size, Doubt Effect and Subjective Beliefs
5.5 Relationship Between Pupillary Size and Veracity by Voting
6 Discussion
7 Limitations and Future Work
8 Conclusion
References
Driver Sleepiness Detection Using LSTM Neural Network
1 Introduction
2 Two Experiments and Their Settings
2.1 Eye-Closure Experiment
2.2 Simulated Driving Experiment
2.3 Data Recording
3 Driver Sleepiness Detection Model
3.1 Visual Marking for Two Alpha-Related Phenomena
3.2 Calculating Alpha Wavelet Energy Threshold Eth
3.3 Detecting Start and End Points of Alpha Waves
3.4 LSTM Classifier
3.5 Using the LSTM Classifier for Classification
4 Experimental Results
4.1 Training Details
4.2 Performance of Detecting Alpha Wave Start and End Points
4.3 Comparison of Three Classifiers
5 Conclusion
References
HTMTAD: A Model to Detect Anomalies of CDN Traffic Based on Improved HTM Network
Abstract
1 Introduction
2 Models and Definitions
2.1 HTM Overview
2.2 Related Concepts
3 Models and Definitions
3.1 Improved Algorithm of Encoder
3.2 Improved Algorithm of Encoder
4 Experiments and Assessment
4.1 Experimental Environment Configuration
4.2 Experimental Results and Analysis
5 Conclusion
Acknowledgements
References
A Deep Learning Based Multi-task Ensemble Model for Intent Detection and Slot Filling in Spoken Language Understanding
1 Introduction
1.1 Problem Definition
1.2 Motivation and Contributions
2 Related Work
3 Methodology
3.1 Embeddings
3.2 Proposed Approach
3.3 Augmenting Syntactic Features
4 Datasets and Experiments
4.1 Datasets
4.2 Training Details
5 Results and Discussion
5.1 Results
5.2 Error Analysis
6 Conclusion and Future Work
References
An Image-Based Approach for Defect Detection on Decorative Sheets
1 Introduction
1.1 Motivation
1.2 Related Work
1.3 Contributions
1.4 Organization
2 Description of Defect Detection
3 Methodology
3.1 Image-Processing-Based Data Augmentation
3.2 Transfer Learning of Deep CNNs
3.3 Multi-model Ensemble
4 Results
4.1 Implementation Details
4.2 Experiments of Individual Models
4.3 Experiments of Multi-model Ensemble
4.4 Test on Realistic Defeat Images
5 Conclusion and Future Work
References
Complex Conditional Generative Adversarial Nets for Multiple Objectives Detection in Aerial Images
Abstract
1 Introduction
2 Methodology
3 Experimental Results
4 Conclusion
Acknowledgements
References
Facial Landmark Detection Under Large Pose
1 Introduction
2 Related Work
2.1 Traditional Methods
2.2 Deep Learning Methods
3 Two-Stage Cascade Regression
3.1 Overview of Our Method
3.2 Tree Based Cascade Regression Model
3.3 Patch-Difference Feature and Feature Selection
3.4 Hard Sample Augmentation
4 Experiments
4.1 Comparison with Other Work
4.2 Incremental Analysis
5 Conclusions
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Neural Information Processing

Description

More details

Other editions

Additional editions

Content

System requirements