Neural Information Processing. Models and Applications

Name: Neural Information Processing. Models and Applications | 17th International Conference, ICONIP 2010, Sydney, Australia, November 21-25, 2010, Proceedings, Part II
Brand: Springer
Price: 96.29 EUR
Availability: OnlineOnly

17th International Conference, ICONIP 2010, Sydney, Australia, November 21-25, 2010, Proceedings, Part II

Kevin K.W. Wong B. Sumudu U. Mendis Abdesselam Bouzerdoum(Herausgeber*in)

Springer (Verlag)

Erschienen am 18. November 2010

XXVI, 740 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-17534-3 (ISBN)

96,29 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Title
Preface
Organization
Table of Contents - Part II
Brain Computer Interface
Utilizing Fuzzy-SVM and a Subject Database to Reduce the Calibration Time of P300-Based BCI
Introduction
Linear Soft Margin Support Vector Machines (C-SVM)
Fuzzy Support Vector Machines (Fuzzy-SVM)
P300 Spellers
Collected Data
Database
Estimation of the C-Parameter and the Weight Values
Results
Conclusion
References
Importance Weighted Extreme Energy Ratio for EEG Classification
Introduction
EER Algorithm: A Brief Review
Importance Weighted EER
Experiment
Data Description and Experimental Setup
Results and Performance Analysis
Conclusion and Future Work
References
Toward Automated Electrode Selection in the Electronic Depth Control Strategy for Multi-unit Recordings
Introduction
Experiments
Inter-rater Reliability
Prediction of Quality Scores
Conclusion
References
Tensor Based Simultaneous Feature Extraction and Sample Weighting for EEG Classification
Introduction
Multilinear Weighted PCA
Matrix Based PCA and Weighted PCA
Multilinear PCA for Tensor Samples
Details of the Proposed Method
Experiment
Conclusions
References
A Tongue-Machine Interface: Detection of Tongue Positions by Glossokinetic Potentials
Introduction
Glossokinetic Potentials
Method
Contralateral Glossokinetic Potentials
Tip for Experiments
Detection of Tongue Positions
Experiments
Conclusions
References
Practical Surface EMG Pattern Classification by Using a Selective Desensitization Neural Network
Introduction
Research Background
Electromyogram
Selective Desensitization Neural Network
Methods
Signal Measurement
Target Hand Movements
Preprocessing of IEMG Signals
Learning of the SDNN
Evaluation of Classification
Experiment
Conclusion
References
Reliability-Based Automatic Repeat reQuest with Error Potential-Based Error Correction for Improving P300 Speller Performance
Introduction
Methods
P300 Speller
Proposed Method
Performance Evaluation
Utility Maximization
Experiment
Data Collection
Data Processing
Results and Discussions
Conclusion
References
An Augmented-Reality Based Brain-Computer Interface for Robot Control
Introduction
Augmented-Reality BCI Design
Experimental Protocol
Methods
Data Acquisition
P300 Classification
Asynchronous Control for P300 BCIs
Experimental Results
Discussion
References
Brain Computer Interfaces: A Recurrent Neural Network Approach
Introduction
Classifiers
Recurrent Neural Network
Echostate Network
CasPer Neural Network
EchoCasPer Neural Network
Pre-processing
Experiment
Conclusion
References
Research on Relationship between Saccade-Related EEG Signals and Selection of Electrode Position by Independent Component Analysis
Introduction
Independent Component Analysis (ICA)
Experimental Settings
Experimental Results
Selection of Electrode
Conclusion
References
Kernel Methods
Application of SVM-Based Filter Using LMS Learning Algorithm for Image Denoising
Introduction
Support Vector Machines
Design of the ASVC Filter
Structure of the ASVC Filter
SVM Impulse Detector
Adaptive Weight of the LUM Filter
Adaptive LUM Filtering
Experimental Results
Conclusion
References
Tuning N-gram String Kernel SVMs via Meta Learning
Introduction
N-gram String Kernel
The Proposed Meta Learning Approach to N-gram SVM Optimization
String Meta-features
Meta Learning on String Categorization
The Proposed Meta Learning Algorithm
Experiments
Datasets
Results on N-gram SVM Optimization
Conclusions and Future Work
References
Bilinear Formulated Multiple Kernel Learning for Multi-class Classification Problem
Introduction
Bilinear Formulation for MC-MKL
Optimization Methods
Optimization with Respect to w
Optimization with Respect to d
Rescaling
Experimental Result
Conclusion
References
Feature Extraction Using Support Vector Machines
Introduction
Kernel Discriminant Analysis
Support Vector Machines
Classification by Support Vector Machines
Feature Extraction by SVM
Computer Experiment
Conclusions
References
Class Information Adapted Kernel for Support Vector Machine
Introduction
Support Vector Machine
Class Informed Kernel
Experiment Setups and Results
Datasets and Software
Experiment with Class Informed Kernel
Varied Support Classifier
SVM as Support Classifier
Conclusion
References
Gaze Pattern and Reading Comprehension
Introduction
Proposed Method
Background
Effective Reduction of Data Resolution
Focus on Back Tracking and Forward Tracking
Levenberg-Marquardt Based Neural Network
Experiment
Background
Setup
Data Collection and Preparation
Evaluation and Comparison
Neural Network Results
Support Vector Machine Comparison
Conclusion
References
A Theoretical Framework for Multi-sphere Support Vector Data Description
Introduction
Support Vector Data Description (SVDD)
One-Class Support Vector Machine (OCSVM)
Multi-sphere SVDD
Problem Formulation
Calculating Radii, Centres and Slack Variables
Calculating Membership U
Iterative Learning Process
Experimental Results
Conclusion
References
Fast Implementation of String-Kernel-Based Support Vector Classifiers by GPU Computing
Introduction
Support Vector Machines and String Kernels
Support Vector Machines
String Kernels
Graphics Processors
Kernels on GPU
p-spectrum Kernel
LD Kernel and Gap-Weighted Subsequence Kernel
Experiment Results
Conclusion and Future Work
References
Model Generation and Classification
Classification of Imbalanced Data by Combining the Complementary Neural Network and SMOTE Algorithm
Introduction
The Proposed Techniques
Complementary Neural Network (CMTNN)
Synthetic Minority Over-Sampling Technique (SMOTE)
The Proposed Combined Techniques
Experiments and Results
Conclusions
References
Generalization Error of Faulty MLPs with Weight Decay Regularizer
Introduction
MLP and Fault Model
Mean Prediction Error of Fault MLPs
Simulations
Conclusion
References
The Effect of Bottlenecks on Generalisation in Backpropagation Neural Networks
Introduction
Algorithm Descriptions
Sensitivity
Distinctiveness
Badness
Adding Noise
Heuristic Pattern Reduction
Bimodal Distribution Removal
Experiments
Method
Results
Conclusions
References
Lagrange Programming Neural Networks for Compressive Sampling
Introduction
Compressive Sampling and LPNN
LPNNs for Compressive Sampling
Sparse Signal
Non-sparse Signal
Measurement Noise
Simulation Results
Artificial Data
Non Sparse Signal
Measurement Noise
Conclusion
References
Input and Output Mapping Sensitive Auto-Associative Multilayer Perceptron for Computer Interface System Based on Image Processing of Laser Pointer Spot
Introduction
Proposed Input-Output Mapping Sensitive AAMLP (I/OSAAMLP)
Laser Pointer Interfacing System Using I/OS-AAMLP
Experiments
Conclusion
References
Improving Recurrent Neural Network Performance Using Transfer Entropy
Introduction
Echo State Networks
The Echo State Property
Transfer Entropy
Reservoir Dynamics
Adaptation of Information Transfer
Experimental Results
Prediction of Autoregressive Coupled Processes
Prediction of Mackey-Glass Time Series
Conclusions
References
Design of Artificial Neural Networks Using Differential Evolution Algorithm
Introduction
Basics on Feed-Forwards Neural Networks
Basics on Differential Evolution
Design of an ANN Using DE
Experimental Results
Conclusions
References
ESNs with One Dimensional Topography
Introduction
Model
Echo State Networks
Design of the Connectivity Matrix
Measuring the Information Transfer in the Reservoir
Results
Discussion
References
Computational Advance in Bioinformatics
iGAPK: Improved GAPK Algorithm for Regulatory DNA Motif Discovery
Introduction
Relative Model Mismatch Score and -Ratio
Relative Model Mismatch Score
a-Ratio
Improved GAPK Algorithm
Search Space Reduction
Evolutionary Computations
Model Refinement
Performance Evaluation
Datasets
Results with Comparisons
Conclusion
References
A Computer-Aided Detection System for Automatic Mammography Mass Identification
Introduction
Methodology
Results
Summary
References
Exploring Features and Classifiers to Classify MicroRNA Expression Profiles of Human Cancer
Introduction
MicroRNA
MicroRNA Expression Classification
Features
Classifiers
Experimental Results
Conclusions
References
SOMIX: Motifs Discovery in Gene Regulatory Sequences Using Self-Organizing Maps
Introduction
System and Methods
Basic Concepts and Problem Formulation
Modeling the Motif Signals
Modeling the Background Signals
Similarity Metric
Hybrid Node Model
Untitled
Algorithm
Adaptation Process
Results
Performance on Real Datasets
Performance on Artificial Datasets with Planted Multiple Motifs
Conclusion
References
Microarray-Based Disease Classification Using Pathway Activities with Negatively Correlated Feature Sets
Introduction
Previous Work
Material and Method
Dataset
Method
Results and Discussion
Conclusions
References
Data Mining for Cybersecurity
A Malware Detection Algorithm Based on Multi-view Fusion
Introduction
Related Work
Multi-view Malware Detection Method Basing on API Call Sequences
Getting API Call Sequences
Characterization of API Call Sequences
Classification Model
Experiment
Experiment Samples
Performance Index
Performance
Conclusion and Future Work
References
A Fast Kernel on Hierarchial Tree Structures and Its Application to Windows Application Behavior Analysis
Introduction
Windows File System and Registry
Kernels on Trees
Kernels on FTrees
Properties of the Kernels
Algorithm
Computational Costs
Experiments
Conclusion and Future Research
References
Evolution of Information Retrieval in Cloud Computing by Redesigning Data Management Architecture from a Scalable Associative Computing Perspective
Introduction
Cloud Data Access Scheme
Efficient Model Using Single-Cycle Learning
Distributed Hierarchical Graph Neuron (DHGN)
Tests and Results
Remarks and Conclusion
References
Factorizing Class Characteristics via Group MEBs Construction
Introduction
Classic MEB
The Proposed g-MEB Learning
Experiments and Discussions
Conclusions and Future Work
References
A Hybrid Fuzzy-Genetic Colour Classification System with Best Colour Space Selection under Dynamically-Changing Illumination
Introduction
Colour Spaces
FCCF-HAGA Algorithm
Fitness Function
Experiment Set-Up
Results and Discussion
Colour Classification Results
Comparison of the Different Colour Classification Algorithms
Conclusions
References
Identifier Based Graph Neuron: A Light Weight Event Classification Scheme for WSN
Introduction
Related Work
System Model
Proposed Scheme
Proposed Protocol
Discussion
Simulation Work
Conclusion and Future Work
References
Clustering Categorical Data Using an Extended Modularity Measure
Introduction
Definitions and Notations
Undirect Graph and Data Matrices
Equivalence Relation
The Relational Analysis Approach
Extensions of the Modularity Measure
Modularity and Graphs
First Extension: Early Integration
Modularity Extensions as a Modified RA Criterion
Optimization Procedure
Modularity Decomposition
Relational Analysis Heuristic
Experiments
Performance Measures
The Datasets for Validation
Results for the Early Integration
Conclusions
References
A Morphological Associative Memory Employing a Reverse Recall
Introduction
Morphological Associative Memory: MAM
Ritter's MAM
MAM Using the Stored Pattern Independent Kernel Image
MAM Employing a Reverse Recall
Experimental Results
Hetero Association Using Twenty-Six Alphabet Characters
Hetero Association Using Sets of Threefold Inclusion Patterns
Conclusion
References
Analysis of Packet Traffics and Detection of Abnormal Traffics Using Pareto Learning Self Organizing Maps
Introduction
Analysis of Packet Traffics Using Pareto Learning Self Organizing Maps
Pareto Learning Self Organizing Maps
Learning IP-Packets Using Occurrence Rate
Experiment of Mapping the Traffics of Packets
Experiments for Detecting Abnormal Packets
Supervised Learning Method
Unsupervised Learning Method
Conclusions
References
Log Analysis of Exploitation in Cloud Computing Environment Using Automated Reasoning
Introduction
Could Computing Environment
Cloud Computing
PaaS, SaaS and IaaS
Virtual Machine Monitor
Mechanized Reasoning
Resolution
Set of Support Strategy
Hyperresolution
Subsumption
Logging Techniques of Windows OS
DLL Injection
Filter Driver
Centralized Logging Architecture
Experimtal Results
Internet Explorer Aurora Attack: MS979352
Experimental Results
Conclusion
References
Self-organizing Maps and Their Applications
A Multidirectional Associative Memory Based on Self-organizing Incremental Neural Network
Introduction
Learning Algorithms
Memory Layer
Associate Layer
Recall and Associate
Recall in Auto-associative Mode
Associate in Hetero-associative Mode
Experiment
Binary (Bipolar) Data
Real Task for Robot with GAM
References
Range Image Registration Using Particle Filter and Competitive Associative Nets
Introduction
Range Image Registration Using PF and CAN2
Modeling of Range Image Registration for the PF
Pair-Wise Registration by the CAN2
Experimental Results
Conclusion
References
Rotation Invariant Categorization of Visual Objects Using Radon Transform and Self-Organizing Modules
Introduction
One SOM Per Direction
The Single-SOM Network
Some Implementation Remarks
Conclusion
References
Learning Topological Constraints in Self-Organizing Map
Introduction
Self-Organizing Maps
Data-Driven Relaxation in Self-Organizing Map
Principle
Algorithm
Experimental Results
Databases Description
Estimation of the Quality of the SOM
Topological Relaxation in SOM
Evaluation of DDR-SOM
Conclusion
References
Pseudo-network Growing for Gradual Interpretation of Input Patterns
Introduction
Theory and Computational Methods
Concept of Pseudo-network Growing
Mutual Information and Information Enhancement
Pseudo-network Growing
Results and Discussion
Conclusion
References
The Adaptive Authentication System for Behavior Biometrics Using Pareto Learning Self Organizing Maps
Introduction
Pareto Learning Self Organizing Maps
Experimental Results of Authentication Using Keystroke Timings and Key Typing Sounds
Experimental Results of Authentication Using Original Data
Experimental Results of Authentication Using the Data Changing by Time
Experimental Results of Authentication Using the Test Data with Variations and Noises
Experimental Results of Incremental Learning of Authenticated User Data
Experimental Results of Detecting Unregistered User Using Threshold
Conclusions
References
Human Action Recognition by SOM Considering the Probability of Spatio-temporal Features
Introduction
3D FAST Corner Detection in ROI
3D FAST Corner Detection
ROI Extraction
Compact Descriptor
3D Gradients and Orientation Quantization
Descriptor Computation
Recognition by SOM
Experiment Results
Conclusion
References
On Generalization Error of Self-Organizing Map
Introduction
Definition of Generalization Error
Distribution of Data
Learning Algorithms of Batch-Learning SOM (BL-SOM)
Generalization Error
Learning Curve
Experiments
Experimental Setting
Experimental Result
Discussion
Conclusions
References
A Novel Approach for Sound Approaching Detection
Introduction
Proposed Method
Experiments
Parameter Decision
Performance Evaluation
Performance Comparison
Conclusions
References
Ground Penetrating Radar System with Integration of Mutimodal Information Based on Mutual Information among Multiple Self-Organizing Maps
Introduction
Integration of Multimodal Information and Concept Generation in the MuSOM
Basic Idea
MuSOM: A Set of SOMs Modulated by Mutual Information
Experiment
Construction of the Adaptive Radar System to Visualize Plastic Landmines
Preprocessing and Feature Extraction
Results
Summary
References
Information-Theoretic Competitive and Cooperative Learning for Self-Organizing Maps
Introduction
Theory and Computational Methods
Two Types of Information
Free Energy Minimization
Competitive and Cooperative Learning
Results and Discussion
Senate Data
Ionosphere Data
Conclusion
References
Early Recognition Based on Co-occurrence of Gesture Patterns
Introduction
Definition of Early Recognition of Gesture Patterns
Typical Gesture Recognition
Early Recognition of Individual Gesture Patterns
Early Recognition of Co-occurring Gesture Patterns
Early Recognition Strategy
System Overview
Sparse Coding
Associative Memory
Similarity Measure
Experimental Results
Condition
Early Recognition Result of Individual Gesture Patterns
Early Recognition Result of Co-occurring Gesture Patterns
Conclusion
References
A Dynamically Reconfigurable Platform for Self-Organizing Neural Network Hardware
Introduction
Self-Organizing Neural Networks
Dynamically Reconfigurable FPGA Platform
Implementation of Self-Organizing Neural Network
Distance Calculation Unit
WTA and Rough WTA Unit
Results and Discussion
Conclusions
References
Inversion of Many-to-one Mappings Using Self-Organising Maps
Bidirectional Backprogration
Normal Backpropagation Training
Reverse Direction Training
Data
dh46
gdp
Interactive Competition Model
Iac Model
Iac Results
Hybrid SOM Model
The Hybrid Model
Input into SOM
Conclusion
References
Self-Organizing Hidden Markov Models
Introduction
EM Algorithm
Self-Organizing Hidden Markov Models
Hidden Markov Models
Estimating HMM Parameters Using the Constrained EM Algorithm
E-step
M-step
State Visualization
Experimental Results
Conclusions
References
An Image-Aided Diagnosis System for Dementia Classification Based on Multiple Features and Self-Organizing Map
Introduction
Flow Chart and Feature Extraction
Spatial Normalization of MRI Data
Volume Features
Shape Features
Self-Organizing Map Architecture
Experimental Results
Material
Statistical Analysis and Classification
Conclusions
References
Parallel Batch Training of the Self-Organizing Map Using OpenCL
Introduction
OpenCL
Implementation of the Batch Training Process Using OpenCL
Finding the BMU
Neuron Adaptation
Results and Discussion
Conclusion
References
Fast Kohonen Feature Map Associative Memory Using Area Representation for Sequential Analog Patterns
Introduction
Fast KFM Associative Memory with Area Representation for Sequential Analog Patterns
Structure
Learning Process
Recall Process
Computer Experiment Results
Association Result
Learning Speed
Storage Capacity
Recall Ability for Sequential Patterns Including Common Terms
Robustness for Damaged Neurons and Noisy Input
Conclusions
References
Machine Learning Applications to Image Analysis
Facial Expression Based Automatic Album Creation
Introduction
Related Work
System
Facial Feature Extraction
Expression Image Formation
Structural Similarity Index
Album Creation and User Options
Empirical Experiment and Outputs
Conclusions and Future Work
References
Age Classification Combining Contour and Texture Feature
Introduction
Age Classification System
Feature Extraction
Contour Feature
Texture Feature
Classifier Combination
Experiments
Conclusions and Future Work
References
A Salient Region Detector for GPU Using a Cellular Automata Architecture
Introduction
Low-Level Architecture for Image Processing
Proposed CA Architecture
Implementation
The Evolutionary Design of Cellular Automata for Image Processing
Training Set and Fitness Function
Mutation Algorithm
Evolved Solution
Performance of CA Salient Region Detectors
Performance Metric
Results
Conclusions
References
VG-RAM WNN Approach to Monocular Depth Perception
Introduction
Visual Cues for Depth Perception
Markov Random Field Approach for Monocular Depth Perception
VG-RAM WNN Approach for Monocular Depth Perception
Methodology
Experiments
Conclusions
References
Semi-supervised Classification by Local Coordination
Introduction
Semi-supervised Local Coordination
MFA for Local Dimension Reduction
Aligning Local Models by Local Coordination
Label Smoothness
Experiment Results
Conclusions
References
RANSAC Based Ellipse Detection with Application to Catadioptric Camera Calibration
Introduction
Ellipse Detection
Central Catadioptric Camera Calibration Based on the Boundary Ellipse
Central Catadioptric Camera
Calibration Based on the Boundary Ellipse
Experimental Results
Conclusion
References
Speed Up Image Annotation Based on LVQ Technique with Affinity Propagation Algorithm
Introduction
Background
Learning Vector Quantization
Affinity Propagation Algorithm
SVM
Proposed Scheme
Training
Annotation
Experimental Results
Experimental Setup
Experiment 1: Preference Parameter Setting for AP
Experiment 2: Selection the Size of Codebook
Experiment 3: Comparative Studies
Conclusion
References
Dictionary of Features in a Biologically Inspired Approach to Image Classification
Introduction
Related Work
Implementation and Results
Clustering over Images from All Classes
Performing Clustering on Images Individually
Clustering over Images of Each Class and Investigating K-NN Algorithm Performance
Sampling over Center of Images
Discussion and Conclusions
References
A Highly Robust Approach Face Recognition Using Hausdorff-Trace Transformation
Introduction
Feature Extraction
The Trace Transform
Robust Identifier
Similarity Measure
The Classical Hausdorff Distance
The Hausdorff- Shape Context
Experimental Results
Conclusions
References
Blind Image Tamper Detection Based on Multimodal Fusion
Introduction
Background
Formulating the Fusion Problem
Experimental Results
Conclusions
References
Orientation Dependence of Surround Modulation in the Population Coding of Figure/Ground
Introduction
The Model of BO Selective Cells
Contrast Detection
Surround Modulation
Results
Discussions
References
Increased Robustness against Background Noise: Pattern Recognition by a Neocognitron
Introduction
Architecture of the Network
Feature-Extracting S-cells
Response of an S-cell
Effect of Noise on S-cells
C-cells: Blur by Root-Mean-Square Operation
Computer Simulation
References
Improving the Performance of Facial Expression Recognition Using Dynamic, Subtle and Regional Features
Introduction
Proposed Approach
Patch-Based Feature Extraction
Patch Matching Operation
Salient' Patch Selection
Experimental Results
Databases
Performance on the JAFFE Database
Performance on the CK Database
Performance with and without Matching Area
Comparison with State-of-the-Art Performance
Conclusion
References
Identity Retrieval in Biometric Access Control Systems Using Multimedia Fusion
Introduction
Multimedia Sensor Fusion
Multimedia Feature Processing
Experimental Results
Conclusions
References
Improvement of Reuse of Classifiers in CBIR Using SVM Active Learning
Introduction
Relevance Feedback
SVM Active Learning for CBIR
Reuse of Classifiers
Improvement of Criterion for Reuse of Classifiers
Experiments
Precision
Leaning Time
Searching Time
Reuse Rate
Conclusions
References
Realizing Hand-Based Biometrics Based on Visible and Infrared Imagery
Introduction
Proposed Solution
Design and Implementation of Acquisition Device
Imagery Pre-processing and Feature Representation
Pre-processing
Feature Extraction
Fusion
Results and Discussion
Verification Performance Using Directional Code
Analysis of Biometric Features
Fusion of Biometric Features
Conclusion
References
Visual Object Detection by Specifying the Scale and Rotation Transformations
Introduction
System Organizations
Size and Rotation Transformation-Specific Similarities
Decomposition of Scale and Rotation
Similarity Computation and Transformation Detection
Simulation Results
Correct Transformation Detection on Facial Images
Detection Performance under Different Conditions
Discussions and Conclusions
References
Multi-view Gender Classification Using Hierarchical Classifiers Structure
Introduction
Hierarchical Classifiers
Angle Categories Selection
Gender Classification Procedure
Experiment
Data Set
Implementation
Result
Complexity Analysis
Conclusions and Future Work
References
Partial Extraction of Edge Filters by Cumulant-Based ICA under Highly Overcomplete Model
Introduction
Model
Linear Overcomplete ICA Model and Partial Estimation
Lower Bound of 4-th Order Covariance in Partial Estimation
Method
Experiments
Conclusion
References
Random Projection Tree and Multiview Embedding for Large-Scale Image Retrieval
Introduction
Random Projection Tree Indexing
Random Projection Tree Construction
Analytical Properties
Multiview Embedding
Experiments
Conclusion
References
Online Gesture Recognition for User Interface on Accelerometer Built-in Mobile Phones
Introduction
Related Work
Gesture Recognition for Online User Interface
Diversity Modeling with Local Templates
Dynamic Time Warping-Based Gesture Recognition
Experimental Result
Gesture-Set Definition and System Implementation
Gesture Recognition Accuracy Test
Template Analysis
Conclusion
References
Constructing Sparse KFDA Using Pre-image Reconstruction
Introduction
KFDA
A Novel Framework for Constructing Sparse KFDA
Basis-Generating-Step
Coefficient-Generating-Step
Experiments
Experimental Settings
Experimental Results
Discussion
Conclusions
References
Applications
Learning Basis Representations of Inverse Dynamics Models for Real-Time Adaptive Control
Introduction
Learning Dynamics for Adaptive Control
Computed Torque Control with Dynamics Model
Learning IDMs from a Training Data Set
Learning Basis Representations of IDMs for Real-Time Adaptive Control
Basis Representation of IDMs
Learning Procedure for PIDM
Real-Time Adaptive Control Method with PIDM
Experiments
Data Generation
Evaluation of the Real-Time Adaptive Control
Conclusion
References
Feel Like an Insect: A Bio-Inspired Tactile Sensor System
Introduction
Previous Work
Sensor Hardware
Interpreting Sensor Readings
Method I: Constraint Based Input Dimension Reduction
Method II: Let the Neural Network Do the Work
Results
Conclusion
References
Spectral Domain Noise Suppression in Dual-Sensor Hyperspectral Imagery Using Gaussian Processes
Introduction
Materials and Methods
Hyperspectral Data
Gaussian Processes for Machine Learning
Absorption Feature Extraction and Parameterization
Results
Effects on Image Spectra
Effects on Parameter Images
Comparison of Parameters from the Image and Spectral Library
Conclusions
References
A High Order Neural Network to Solve Crossbar Switch Problem
Introduction
High-Order Discrete Hopfield Network
Crossbar Switch Problem
Problem Description
Constructing Energy Function of Crossbar Switch Problem
Constructing High Order Hopfield Neural Network
Strategy for Escaping from Local Minima
Simulation Results
Summary
References
Identification of Liquid State of Scrap in Electric Arc Furnace by the Use of Computational Intelligence Methods
Formulation of the Problem
Signal Preprocessing
Estimation of Variance with Sliding Window
Nonparametric Noise Estimation
Analysis of Symmetry between Signals
Experiments and Results
Dataset Description
Test Procedure
Results
Discussion of the Obtained Results and Conclusions
References
Simulating Wheat Yield in New South Wales of Australia Using Interpolation and Neural Networks
Introduction
Data Pre-processing and Statistical Analysis
Expansion of Training Data Using 2D Interpolations
Simulating Wheat Yield Using Neural Networks
Discussion and Conclusion
References
Investment Appraisal under Uncertainty - A Fuzzy Real Options Approach
Introduction
The Valuation Approach
Illustrative Examples
Option to Defer
Option to Abandon
Sequential Multiple Options
Discussion and Conclusion
References
Developing a Robust Prediction Interval Based Criterion for Neural Network Model Selection
Introduction
Theory and Background
Proposed Model Selection Criterion
Simulation Results and Discussion
Synthetic Case Study
Real Case Study
Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken