Partially Supervised Learning
First IAPR TC3 Workshop, PSL 2011, Ulm, Germany, September 15-16, 2011, Revised Selected Papers
Published on 9. February 2012
X, 159 pages
978-3-642-28258-4 (ISBN)
Description
This book constitutes thoroughly refereed revised selected papers from the First IAPR TC3 Workshop on Partially Supervised Learning, PSL 2011, held in Ulm, Germany, in September 2011.
The 14 papers presented in this volume were carefully reviewed and selected for inclusion in the book, which also includes 3 invited talks. PSL 2011 dealt with methodological issues as well as real-world applications of PSL. The main methodological issues were: combination of supervised and unsupervised learning; diffusion learning; semi-supervised classification, regression, and clustering; learning with deep architectures; active learning; PSL with vague, fuzzy, or uncertain teaching signals; learning, or statistical pattern recognition; and PSL in cognitive systems. Applications of PSL included: image and signal processing; multi-modal information processing; sensor/information fusion; human computer interaction; data mining and Web mining; forensic anthropology; and bioinformatics.
More details
Other editions
Additional editions
Friedhelm Schwenker | Edmondo Trentin
Partially Supervised Learning
First IAPR TC3 Workshop, PSL 2011, Ulm, Germany, September 15-16, 2011, Revised Selected Papers
Book
02/2012
1st Edition
Springer
€53.49
Shipment within 7-9 days
Content
- Title page
- Preface
- Organization
- Table of Contents
- Invited Talks
- Unlabeled Data and Multiple Views
- Introduction
- Semi-supervised Learning and Multi-view
- Active Learning and Multi-view
- About the Views
- Conclusion
- References
- Online Semi-supervised Ensemble Updates for fMRI Data
- Introduction
- Methods
- Online Linear Discriminant Classifier
- Naive Labelling
- Random Subspace Ensemble
- Guided Update Strategy
- Theory and Simulations
- Experiment with fMRI Data
- Dataset: Bangor 1
- Protocol
- Results
- Conclusion
- References
- Studying Self- and Active-Training Methods for Multi-feature Set Emotion Recognition
- Introduction
- Methods
- Supervised Learning Experiment
- Self-training Experiment
- Active Learning Experiment
- Dataset Description
- Features
- Experiments
- Supervised Learning
- Self-training Experiment
- Active Learning Experiment
- Discussion
- Conclusions
- References
- Algorithms
- Semi-supervised Linear Discriminant Analysis Using Moment Constraints
- Introduction
- Related Work
- Semi-supervised Nearest Means
- Semi-supervised LDA
- Experimental Setup and Results
- Discussion and Conclusion
- References
- Manifold-Regularized Minimax Probability Machine
- Introduction
- Related Work
- Minimax Probability Machine
- Manifold Regularization
- Semi-supervised Minimax Probability Machine
- Manifold-Regularized Minimax Probability Machine
- Algorithm
- Kernelization
- Experiment
- Discussion
- References
- Supervised and Unsupervised Co-training of Adaptive Activation Functions in Neural Nets
- Introduction
- The Training Algorithm
- Generation of Locally-Supervised Training Sets
- Probabilistic Weighting of Patterns
- Partially Supervised Maximum-Likelihood Estimation of the Probabilistic Weights
- Preliminary Conclusions
- References
- Semi-unsupervised Weighted Maximum-Likelihood Estimation of Joint Densities for the Co-training of Adaptive Activation Functions
- Introduction
- Maximum Likelihood Estimation with Weighted Patterns
- Extension to Multiple Hidden Layers
- Demonstration
- Conclusion
- References
- Semi-Supervised Kernel Clustering with Sample-to-Cluster Weights
- Introduction
- Sample-to-Cluster Weighted Error Function
- Weighted Kernel Based Methods for Clustering
- Semi-Supervised Kernel Clustering with Sample-to-Cluster Weights
- Experiments and Results
- Steel Plates Faults Dataset
- MiniBooNE Dataset
- Conclusion
- References
- Homeokinetic Reinforcement Learning
- Introduction
- Reinforcement Learning in Continuous Space and Time
- Learning in Motor Space
- Homeokinetic Reinforcement Learning: Experiments
- Performance in a Toy Example
- Self-organisation of Walking in an Hexapod
- Results and Discussion
- Pendulum Results
- Hexapod Results
- Conclusion
- References
- Iterative Refinement of HMM and HCRF for Sequence Classification
- Introduction
- Iterative Refinement of HMM and HCRF
- Experiments
- Conclusion
- References
- Applications
- On the Utility of Partially Labeled Data for Classification of Microarray Data
- Introduction
- Methods
- Supervised Learning
- Semi-supervised Learning
- l×k Cross-Validation
- Cross-Validation Experiments for Semi-supervised Classifiers
- Algorithms
- Experimental Setup
- Results
- Discussion
- References
- Multi-instance Methods for Partially Supervised Image Segmentation
- Introduction
- Related Work
- Proposal Object Segments
- Multi-instance Methods
- Semantic Scene Segmentation via Multi-instance Learning
- Multi-instance Kernels for Image Segmentation
- Constraint Parametric Min Cuts (CPMC)
- Multi-instance Learning Using MI-Kernels
- Segment Features
- Combining Segments
- Experiments
- Instance-Level Predictions Using Multi-instance Kernels
- Partially Supervised Image Segmentation on Graz 02
- Conclusions
- References
- Semi-supervised Training Set Adaption to Unknown Countries for Traffic Sign Classifiers
- Introduction
- Related Work and Applied Methods
- Prerequisites
- Traffic Sign Recognition
- Classification
- Self-learning
- Sample Selection
- Confidence Bands
- Virtual Training Samples
- Experimental Evaluation
- Experimental Setup
- Self-learning with Real Traffic Signs
- Self-learning with Virtual Traffic Signs
- Summary and Conclusions
- References
- Comparison of Combined Probabilistic Connectionist Models in a Forensic Application
- Introduction
- Probabilistic Interpretation of Neural Networks
- Combination Techniques
- Experiments
- Conclusions
- References
- Classification of Emotional States in a Woz Scenario Exploiting Labeled and Unlabeled Bio-physiological Data
- Introduction and Related Work
- Data Collection
- Problem Statement and Proposed Method
- Experiments and Results
- Discussion and Future Work
- References
- Using Self Organizing Maps to Find Good Comparison Universities
- Introduction
- Background
- Methods
- Results
- Discussion
- Conclusions
- References
- Sink Web Pages in Web Application
- Introduction
- Defining a Relation between Two Web Pages
- The Concept of Sink Web Page
- Conclusion and Future Work
- References
- Author Index
