Artificial Neural Networks and Machine Learning - ICANN 2011

Name: Artificial Neural Networks and Machine Learning - ICANN 2011 | 21st International Conference on Artificial Neural Networks, Espoo, Finland, June 14-17, 2011, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

21st International Conference on Artificial Neural Networks, Espoo, Finland, June 14-17, 2011, Proceedings, Part II

Timo Honkela Wlodzislaw Duch Mark Girolami Samuel Kaski(Editor)

Springer (Publisher)

Published on 13. June 2011

XXII, 474 pages

E-Book

PDF with digital watermarking

System requirements

978-3-642-21738-8 (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

Title
Preface
Organization
Table of Contents
A Markov Random Field Approach to Neural Encoding and Decoding
Introduction
Methods
Encoding
Decoding
Experiments
Stimuli, Acquisition and Data Analysis
Encoding Analysis
Decoding Analysis
Conclusions
References
Weakly Supervised Learning of Foreground-Background Segmentation Using Masked RBMs
Introduction
Model
Experiments
Results
Discussion
References
Recursive Multi-Way PLS for Adaptive Calibration of Brain Computer Interface System
Introduction
Methods
Generic PLS
Recursive PLS
NPLS
RNPLS
Results
References
Transformation of Edge Weights in a Graph Bipartitioning Problem
Introduction
The Double-Step Algorithm
Obtained Results
Conclusions
References
A Distributed Behavioral Model Using Neural Fields
Introduction
Behavioral Model
Stimulus Design
Implementation and Results
Conclusion
References
A Hypothetical Free Synaptic Energy Function and Related States of Synchrony
Introduction
Model
TheAnsatz
Analysis and Results
Discussion
Conclusion
References
Observational Learning Based on Models of Overlapping Pathways
Introduction
Computational Modeling
Input Encoding
Neurons and Connections
Model Pathways
Motor Control of the Simulated Agent
Results
Discussion
References
On the Capacity of Transient Internal States in Liquid-State Machines
Introduction
Liquid-State Machines and Internal States
Simulation Results and Discussion
Conclusion
References
Hybrid Parallel Classifiers for Semantic Subspace Learning
Introduction
Methodology Overview and Overall Architecture
Data Vector Sets Generation
Classification Algorithms
Results and Their Analysis
Conclusion
References
Temperature Prediction in Electric Arc Furnace with Neural Network Tree
Introduction
Method
Experimental Results
Conclusions
References
Optimizing Linear Discriminant Error Correcting Output Codes Using Particle Swarm Optimization
Introduction
Linear Discriminant Error Correcting Output Codes Framework
Particle Swarm Optimization
Experimental Results
Conclusion
References
SOS-HMM: Self-Organizing Structure of Hidden Markov Model
Introduction
From Self-Organizing Map to SOS-HMM
Experimentations
Conclusion
References
Image Receptive Fields Neural Networks for Object Recognition
Introduction
Image Receptive Fields Neural Network
Structure of the Network
Image and Receptive Fields Neurons
Geometrical Figures in Binary Images
Presentation of the Experiments
Main Results
Photographic Object Recognition - the ALOI Database
Conclusions and Perspectives
References
A Comparison of the Electric Potential through the Membranes of Ganglion Neurons and Neuroblastoma Cells
Introduction
The Membrane Model
The Electric Potential in the Membrane Regions
Surface Potentials
Results
Conclusions
References
SNPboost: Interaction Analysis and Risk Prediction on GWA Data
Introduction
Methods and Data
Support Vector Machine (SVM)
Adaboost
SNPboost: Adaboost for GWA Data
Data
Results and Discussion
Linear Classification
Non-linear Classification
Interaction Analysis
Conclusion
References
Binary Patterns Identification by Vector Neural Network with Measure of Proximity between Neuron States
Introduction
Model Description
Binary Vector Pre-processing
Efficiency of Identification
Conclusion
References
Emerging Bayesian Priors in a Self-Organizing Recurrent Network
Introduction
Self-Organizing Recurrent Network (SORN)
Network Architecture
Plasticity Mechanisms
Spontaneous Activity in SORNs Matches the Statistics of the Inputs
Influence of Learned Bayesian Priors on Classification
Bistable Stimuli
Multiple Sources of Information
Prediction Cues
Conclusions
References
Momentum Acceleration of Least-Squares Support Vector Machines
Introduction
Algorithms for LS-SVM Training
Momentum Acceleration
Experiments
Discussion and Further Work
References
Fast Support Vector Training by Newton's Method
Introduction
Support Vector Machines
Training Methods
Calculating Corrections by Newton's Method
Working Set Selection
Calculating Corrections by Cholesky Factorization
Detection of Two-Cycle and One-Cycle Loops
Performance Evaluation
Conclusions
References
Linear Operators and Stochastic Partial Differential Equations in Gaussian Process Regression
Introduction
Linear Operators and Functionals of GPs
GP Regression Model with Linear Operators
Stochastic Partial Differential Equations
Simulation: Electrostatic Inverse Problem
Conclusion
References
Learning from Multiple Annotators with Gaussian Processes
Introduction
Gaussian Process Regression
Multi-annotator Regression
Experiments
Conclusions and Future Work
References
Estimation of the Number of Clusters Using Heterogeneous Multiple Classifier System
Introduction
Heterogeneous Ensemble Classifier
Clustering Ensemble Generation
Detection of the Ambiguous Patterns
Estimation of the Number of Clusters
Re-clustering of Ambiguous Patterns
Experimental Evaluation
Conclusion
References
A Distributed Self-adaptive Nonparametric Change-Detection Test for Sensor/Actuator Networks
Introduction
Problem Statement
Distributed Change-Detection Test
ICI-Based Change-Detection Test at the Unit Level
The Change-Detection Test at the Data-Aggregation Center
Experiments
Conclusions
References
Weighted Mutual Information for Feature Selection
Introduction
Weighted Mutual Information
Feature Selection with Weighted Mutual Information
Related Work
Experiments
Conclusion
References
Face Prediction from fMRI Data during Movie Stimulus: Strategies for Feature Selection
Introduction
Methods and Materials
Regression Model and Feature Selection
Functional MRI Experiment
Results
Conclusion
References
The Authentication System for Multi-modal Behavior Biometrics Using Concurrent Pareto Learning SOM
Introduction
Pareto Learning SOM
Pareto Learning SOM for Multi Modal Vector
Concurrent Pareto learnig SOM
Experimental Results
Biometric Authentication Using Keystroke Timings and Key Typing Sound
Experimental Result Using CP-SOM
Conclusion
References
Hermite Polynomials and Measures of Non-gaussianity
Introduction
The Hermite Polynomial Series
Definition
Convergence of the Series
Measures of Non-gaussianity
Definition
Simulations
Conclusions
References
Complex-Valued Independent Component Analysis of Natural Images
Introduction
Complex Independent Component Analysis and Its Limitations
Simulations with Natural Images
Checking Model Assumptions
Extension of Complex ICA
Conclusions
References
Improving Gaussian Process Value Function Approximation in Policy Gradient Algorithms
Introduction
Estimating Value Functions with Gpr
Redundancy in Value Function Approximation
Maintaining Continuity
Experiments
Conclusions and Further Research
References
Application of Nonlinear Neural Network Model for Self Sensing Characteristic in an Ionic Polymer Metal Composite (IPMC) Actuator
Introduction
Experimental Apparatus
Nonlinear Black-Box Model for the IPMC Actuator Based on RMLPNN and SALM
Multi-Layer Perceptions Neural Network (MLPNN) Structure
Nonlinear Black-Box Model for IPMC Actuator
Modeling Results
Conclusion
References
Optimal Control Using Functional Type SIRMs Fuzzy Reasoning Method
Introduction
SIRMs Fuzzy Model
Nonlinear Fuzzy Feedback Control
Functional Analytic Properties of Fuzzy Model
The Set of Membership Functions and Its Properties
Admissible Fuzzy Controller
Lipschitz Continuity and Unique Solution of State Equation
Application to Optimal Control Problem
Conclusion
References
High-Dimensional Surveillance
Introduction
Background for Surveillance
Transform to Supervised Learning
Examples
High-Dimensional Data
Data with Multiple Locations
Disease Clusters with Attribute Interaction
Conclusions
References
A One-Layer Dual Recurrent Neural Network with a Heaviside Step Activation Function for Linear Programming with Its Linear Assignment Application
Introduction
Model Description
Convergence and Optimality Analysis
Linear Assignment Application
Simulation Results
Conclusions
References
Neural Network Solution of Optimal Control Problem with Control and State Constraints
Introduction
Optimal Control Problem with Control and State Constraints
Neural Network Solution of Optimal Control Problem
Model of Feeding Adaptation and Neural Network Simulation
Conclusion
References
Singular Perturbation Approach with Matsuoka Oscillator and Synchronization Phenomena
Introduction
Coupled Systems of Matsuoka Oscillators
Singular Perturbation Approach
Simulation Results
Discussion
References
A RANSAC-Based ISOMAP for Filiform Manifolds in Nonlinear Dynamical Systems
Introduction
Mass-Spring Model for Chaotic Dripping Faucet
Analysis
Application of the Conventional ISOMAP
RANASC-Based ISOMAP and Its Results
Summary and Discussion
References
Manifold Learning for Visualization of Vibrational States of a Rotating Machine
Introduction
Low Dimensional Representation of Time Series Dynamics
Feature Extraction
Dimensionality Reduction
Experiments
Results
Conclusions
References
Bias of Importance Measures for Multi-valued Attributes and Solutions
Introduction
Attribute Importance Measures
Attribute Importance from OOBForest and pForest
Experiments
Conclusion
References
A Computationally Efficient Information Estimator for Weighted Data
Introduction
Information Estimator
Information Estimator Based on Quantiles
Mean Quantile Information Estimator
Entropy Estimator Based on Quantiles
Numerical Experiments
Conclusion
References
Top-Down Induction of Reduced Ordered Decision Diagrams from Neural Networks
Introduction
Another View on Rule Extraction
Reduced Ordered Decision Diagrams
Description of the Algorithm
Experimental Results
Conclusion
References
A Framework for Application-Oriented Design of Large-Scale Neural Networks
Introduction
Pattern-Based Networks
A C++ Framework for Network Construction
Population model
Area Templates
Patch Template
Pattern Storage
A Simpler Description
Conclusion and Future Work
References
A Dynamic Field Model of Ordinal and Timing Properties of Sequential Events
Introduction
Model Description
Model Equations
Simulations Results
Discussion
References
Robot Trajectory Prediction and Recognition Based on a Computational Mirror Neurons Model
Introduction
SRNPB for Robot Trajectory Recognition
Experimental Results
Conclusion and Future Works
References
A Sentence Generation Network That Learns Surface and Abstract Syntactic Structures
Introduction
Meanings Represented as Sensorimotor Sequences
Architecture
The Episode Rehearsal Network
The Word Production Network
The Control Network
Learning Surface Patterns
The Entropy Network
Sentence Generation in the Trained System
Training the System
Experiment
Results
Conclusion
References
A Perceptual Memory System for Affordance Learning in Humanoid Robots
Introduction
Memory as a Basis for Learning in Cognitive Robotics
Our Perceptual Memory System
Experimental Results
Conclusion and Outlook
References
Probabilistic Proactive Timeline Browser
Introduction
Timeline Browser with a Relevance Estimator
Relevance Prediction
Experiments
Comparing the Interfaces
Integrating Explicit and Implicit Feedback
Discussion
References
Person Tracking Based on a Hybrid Neural Probabilistic Model
Introduction
Methods
Particle Filter
Sigma-Pi Network
Processing Different Visual Cues
Experimental Results
Tracking a Person
Changing Environment
Discussion
Conclusions
References
Gaze- and Speech-Enhanced Content-Based Image Retrieval in Image Tagging
Introduction
Gaze-Enhanced Content-Based Image Retrieval
PicSOM CBIR System
Using Gaze Patterns as Implicit Relevance Feedback
CBIR-Assisted Image Tag Correction
Automatic Speech Recognition
Experiments
Conclusions
References
Modelling Hypothetical Wage Equation by Neural Networks
Introduction
Description of Used Methods
Unit Root Tests
Multilinear Regression
Feed-Forward Back-Propagation Neural Network
Experiments
Variable Selection
Modelling
Results
Conclusions
References
On the Designing of Spikes Band-Pass Filters for FPGA
Introduction
Building Blocks for SSP
Spike Synthetic Generator (SSG)
Spikes Integrate and Generate (SI&G)
Spikes Hold and Fire (SH&F)
Spike Frequency Divider (SFD)
Spike Low-Pass Filters (SLPF)
Low Q-Factor Spikes Band-Pass Filter (LQ-SBPF)
High Q-Factor Spike Band-Pass Filter (HQ-BPF)
Conclusions
References
An Information Geometrical View of Stationary Subspace Analysis
Introduction
Block Gaussian Model for SSA
Maximum Likelihood and Its Information Geometric Interpretation
Numerical Example
Conclusions
References
Forecasting Road Condition after Maintenance Works by Linear Methods and Radial Basis Function Networks
Introduction
Description of the Road Data
Formulation of the Forecasting Problem
Forecasting Methods
Current Methods in Finnish Transport Agency
Sequential Input Selection Algorithm (SISAL)
Radial Basis Function (RBF) Networks
Experiments and Results
Summary and Conclusions
References
Multistart Strategy Using Delta Test for Variable Selection
Introduction
Delta Test
Forward-Backward Selection
Multistart/Restart Strategies
Multistart Strategy for Delta Test
Experiments
Synthetic Data
Real-World Data
Conclusions
References
Speech Recognition Based on the Processing Solutions of Auditory Cortex
Introduction
Methods
Model Dynamics and Structure
Stimuli
Analysis
Results
Discussion
References
A Geometric Bio-inspired Model for Recognition of Low-Level Structures
Introduction
Atomic Functions
Geometric Algebra
Rotors in G3
Quaternion Algebra
Quaternion Atomic Function Qup
Geometric Radial Basis Function Networks
Experimental Analysis
Conclusions
References
View-Tuned Approximate Partial Matching Kernel from Hierarchical Growing Neural Gases
Introduction
Related Work
Approximate Partial Matching with lbTreeGNGs
View-Tuned Approximate Partial Matching
Conclusion and Future Work
References
ANGE - Automatic Neural Generator
Introduction
Neuron Implementation
ANGE Details
Test Results
Conclusion
Further work
References
Uncertainty Sampling-Based Active Selection of Datasetoids for Meta-learning
Introduction
Meta-learning for Algorithm Selection
Active Learning and Meta-Learning
Experiments and Results
Conclusion
References
Learning Scheme for Complex Neural Networks Using Simultaneous Perturbation
Introduction
Complex-Valued Neural Networks and the Simultaneous Perturbation Method
Complex Back-Propagation Learning Rule
Simultaneous Perturbation Learning Rule for Complex-Valued Neural Networks
Comparison between Simultaneous Perturbation and Back-Propagation
Exclusive OR Problem
Scaling Problem
Rotation Problem
Conclusion
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Artificial Neural Networks and Machine Learning - ICANN 2011

Description

More details

Other editions

Additional editions

Content

System requirements