Computational Intelligence: Research Frontiers: IEEE World Congress on Computational Intelligence, WCCI 2008, Hong Kong, China, June 1-6, 2008, ... (Lecture Notes in Computer Science, 5050)

The 2008 IEEE World Congress on Computational Intelligence (WCCI 2008), held during June 1–6, 2008 in Hong Kong, China, marked an important milestone in advancing the paradigms of the new fields of computational intelligence. As the fifth event in the series that has spanned the globe (Orlando-1994, Anchorage-1998, Honolulu-2002, Vancouver-2006), the congress offered renewed and refreshing focus on the progress in nature-inspired and linguistically motivated computation. Most of the congress’s program featured regular and special technical sessions that provided participants with new insights into the most recent developments in the field. As a tradition, in addition to the parallel technical sessions, WCCI holds a series of plenary and invited lectures which are not included in the congress proceedings. As its predecessors, at WCCI 2008, 20 expert speakers shared their expertise on broader, if not panoramic, topics spanning a diverse spectrum of computational intelligence in the areas of neurocomputing, fuzzy systems, evolutionary computation, and adjacent areas. Thanks to their time and expertise, we endeavored to offer this volume to attendees directly at the congress and the general public afterwards. Title Page Preface Organisation Table of Contents A New Framework for Machine Learning Introduction Bayesian Methods Graphical Models Approximate Inference Example Application: Bayesian Ranking Discussion Bilevel Optimization and Machine Learning Introduction Bilevel Optimization A Bilevel Support-Vector Regression Model Bilevel Problems as MPECs Alternative Bilevel Optimization Methods A Relaxed NLP Reformulation Penalty Reformulation Successive Linearization Algorithm for Model Selection Early Stopping Grid Search Experimental Design Synthetic Data Real-World QSAR Data Post-processing Computational Results: Synthetic Data Computational Results: QSAR Data Discussion Bayesian Ying Yang System, Best Harmony Learning, and Gaussian Manifold Based Family Introduction Two Intelligent Abilities and Three Inverse Problems Efforts Towards Challenges Two-Pathway Approaches and the Scope of This Paper Bayesian Ying-Yang Learning Bayesian Ying-Yang System and Best Harmony Learning Yang Machine: Implementable Scenarios Ying Machine : Distributed Log-Quadratic Inner Structures Best Harmony vs Best Matching: Relations to Others Special Cases: Relations to Existing Approaches Best Harmony Versus Best Matching BKYY Learning, Helmholtz Machine, and Variational Approach A Relationship Map Gaussian Manifold Based Systems, Typical Applications, and Concluding Remarks The Berlin Brain-Computer Interface Introduction The Machine Learning Approach Neurophysiological Features Processing and Machine Learning Techniques Common Spatial Patterns Analysis Regularized Linear Classification BBCI Control Using Motor Paradigms High Information Transfer Rates Good Performance without Subject Training Automatic Response Verification Applications of BBCI Technology Prosthetic Control Time-Critical Applications: Prediction of Upcoming Movements Neuro Usability Mental State Monitoring Conclusion Basic Scheme of Neuroinformatics Platform: XooNIps Introduction XooNIps Main View System Outline XooNIps Features and Benefits System Architecture XooNIps Based VP and Other Application The Other Applications of XooNIps Platforms under Japan-Node Japan-Node Access Statistics of the Japan-Node Portal Overview of NI-Platforms under Japan-Node Conclusion Collaborative Architectures of Fuzzy Modeling Introductory Comments Fuzzy Clustering, Information Granules and Communication Mechanisms Collaborative Clustering The General Flow of Collaborative Processing Evaluation of the Quality of Collaboration: Forming a Compromise between Global and Local Characteristics of Data Fuzzy Sets of Type-2 in the Quantification of the Effect of Collaboration Hierarchical Clusters of Clusters Experience Consistent Fuzzy Models: A Concept The Experience-Consistent Development of the Rule-Based Model The Construction of Information Granules of Conditions of the Rules The Consistency-Based Optimization of Local Regression Models The Alignment of Information Granules Characterization of Experience-Consistent Models through Its Granular Parameters Conclusions References Information Fusion for Man-Machine Cooperation Introduction Variables and Question Answering Basic Knowledge Representation Using Fuzzy Sets On the Measures of Possibility and Certainty Hedging on Our Data Multi-source Information Fusion Multiple Fused Values from Multi-source Data Fusing Probabilistic and Possibilistic Data Conclusion References Bio-inspired Self-Organizing Relationship Network as Knowledge Acquisition Tool and Fuzzy Inference Engine Introduction Self-Organizing Relationship (SOR) Network Learning Mode of the SOR Network Execution Mode of the SOR Network Trailer-Truck Back-Up Control Acquisition of Learning Vectors and Their Evaluation by Fuzzy Inference Computer Simulation I/O Characteristics of SOR Network Trajectories of Trailer-Truck Effectiveness of Repulsive Learning Practical Experiment Conclusions References Type-2 Fuzzy Logic Controllers: A Way Forward for Fuzzy Systems in Real World Environments Introduction Type-2 Fuzzy Sets Interval Type-2 FLC Avoiding the Computational Overheads of Type-2 FLCs Type-Reduction Approximation Type-2 Hierarchical Fuzzy Logic Controllers Hardware Implementations and Type-2 Co-processors Successful Applications of Type-2 FLCs Applications of Type-2 FLCs to Industrial Control Applications of Type-2 FLCs to Robot Control The Application of Type-2 FLCs to Ambient Intelligent Environments Control Conclusions References The Burden of Proof: Part II Introduction K-Armed Bandits and Minimizing Expected Losses Methods on a 2-Armed Bandit Problem Results on the 2-Armed Bandit Simulation Discussion on Minimizing Expected Losses Evolutionary Unstable Strategies Background on the Hawk-Dove Game Experimental Results Showing ESSs Are Not Stable Discussion of Evolutionary Unstable Strategies Expecting the Unexpected at the El Farol The El Farol Problem Experimental Methods Results of Evolving Predictors for the El Farol Problem Discussion on El Farol Results Conclusions References Evolution of Altruistic Robots Altruistic Cooperation in Nature Artificial Evolution of Cooperation Evolutionary Conditions Altruistic Foraging Altruistic Communication Conclusion Simulated Evolution under Multiple Criteria Conditions Revisited Introduction Characteristics of Multiobjective Optimization Evolutionary Multiobjective Optimization (EMO) Individual-Based Approaches Population-Based Approaches More Bio-inspired Algorithms Diploid Genomes Multicellular Individuals Gene Duplication and Gene Deletion Gender Dimorphism Predators and Prey Conclusions Handling Uncertainties in Evolutionary Multi-Objective Optimization Introduction Background Information Noisy MO Optimization Handling Noisy Multi-Objective Optimization Multi-Objective Evolutionary Algorithm with Robust Features Simulation Results Dynamic Multi-Objective Optimization Handling Dynamic Multi-Objective Optimization Competitive-Cooperative Coevolution for Dynamic Multi-Objective Optimization Competitive-Cooperative Coevolution. Introducing Diversity via Stochastic Competitors. Handling Outdated Archived Solutions. Simulation Results Robust Multi-Objective Optimization Robust Multi-Objective Problem Test Suite Evolutionary Robust Optimization Techniques Single-objective approach. Multi-objective approach. Solving Vehicle Routing Problem with Stochastic Demand Local Search Heuristic. Route Simulation Method. Simulation Results. Conclusion VCV2 – Visual Cluster Validity Introduction The VAT Image Transforming the Partition Matrix Numerical Examples Discussion and Conclusions References Data Management by Self-Organizing Maps Introduction The Classical Vector Quantization (VQ) The Self-Organizing Map (SOM) General Calibration of the SOM Comparison and Classification of Input Items on the Basis of Features Main Application Areas of the SOM Learning Principles of the SOM General The Original, Stepwise Recursive SOM Algorithm The Batch Computation of the SOM Applications of the SOM Ordering of Countries on the Basis of Sets of Socioeconomic Indicators SOMs of Very Large Document Collections Approximation of an Input Data Item by a Linear Mixture of Models Fitting with the Nonnegativity Constraint The lsqnonneg Function Description of a Document by a Linear Mixture of SOM Models Discussion Cocktail Party Processing Introduction Computational Auditory Scene Analysis Peripheral Analysis and Feature Extraction Auditory Segmentation Voiced Speech Segregation Unvoiced Speech Segregation Discussion Similarities in Fuzzy Data Mining: From a Cognitive View to Real-World Applications Introduction Similarity and Categorization in Cognitive Science Categorization Similarity Related Concepts Similarities in Data Mining Standard Data Mining Similarities in Fuzzy Data Mining Similarities in a Fuzzy Setting Measures of Similarity General Framework for Measures of Comparison Properties of Measures of Comparison Similarity-Based Prototypes Examples of Utilization in Real Word Applications Image Interpretation Defect Forecasting Risk Rating Web Usage Mining Content-Based Image Retrieval Conclusion References Attaining Fault Tolerance through Self-adaption: The Strengths and Weaknesses of Evolvable Hardware Approaches Introduction Background Basics of Evolvable Hardware Real-Time Systems What Is Fault Tolerance? Some Common Myths Putting Fault Tolerance into Practice Fault Detection Methods Fault Recovery Methods Fault Masking How to Design a Fault Tolerant System How Evolvable Hardware Supports Fault Recovery Conclusions Author Index

this State-of-the-art Survey Offers A Renewed And Refreshing Focus On The Progress In Nature-inspired And Linguistically Motivated Computation. The Book Presents The Expertise And Experiences Of Leading Researchers Spanning A Diverse Spectrum Of Computational Intelligence In The Areas Of Neurocomputing, Fuzzy Systems, Evolutionary Computation, And Adjacent Areas. The Result Is A Balanced Contribution To The Field Of Computational Intelligence That Should Serve The Community Not Only As A Survey And A Reference, But Also As An Inspiration For The Future Advancement Of The State Of The Art Of The Field.

the 18 Selected Chapters Originate From Lectures And Presentations Given At The 5th Ieee World Congress On Computational Intelligence, Wcci 2008, Held In Hong Kong, China, In June 2008. After An Introduction To The Field And An Overview Of The Volume, The Chapters Are Divided Into Four Topical Sections On Machine Learning And Brain Computer Interface, Fuzzy Modeling And Control, Computational Evolution, And Applications.

This state-of-the-art survey offers a renewed and refreshing focus on the progress in nature-inspired and linguistically motivated computation. The book presents the expertise and experiences of leading researchers spanning a diverse spectrum of computational intelligence in the areas of neurocomputing, fuzzy systems, evolutionary computation, and adjacent areas. The result is a balanced contribution to the field of computational intelligence that should serve the community not only as a survey and a reference, but also as an inspiration for the future advancement of the state of the art of the field. The 18 selected chapters originate from lectures and presentations given at the 5th IEEE World Congress on Computational Intelligence, WCCI 2008, held in Hong Kong, China, in June 2008. After an introduction to the field and an overview of the volume, the chapters are divided into four topical sections on machine learning and brain computer interface, fuzzy modeling and control, computational evolution, and applications