Inhalt

•  
   Contents
•  
   Introduction
•  
   Contributions and Manuscript Structure
•  
   Publications in the Context of this Thesis
•  
  Historical Background
•  
   Biologically Inspired Learning
•  
   Pseudo-Incremental Learning
•  
   Early Incremental/Online Learning
•  
   Support Vector Machine and Convex Optimization
•  
   The Rise of Tree Ensembles
•  
   Current State
•  
   Incremental Learning
•  
  Overarching Learning Scenario
•  
   Definition
•  
   Challenges
•  
   Incremental Learning Vector Quantization
•  
   Foundation
•  
   Related Work
•  
   Learning Architecture
•  
   Proposed Placement Strategy: COSMOS
•  
   Experiments
•  
   Discussion
•  
   Local Split-Time Prediction
•  
   Foundation
•  
   Related Work
•  
   Proposed Method: OSM
•  
   Experiments
•  
   Discussion
•  
   A Practice-Oriented Survey
•  
   Foundation
•  
   Related Work
•  
   Datasets and Implementations
•  
   Hyperparameter Optimization
•  
   Measure of Model Complexity
•  
   Evaluation Settings
•  
   Experiments
•  
   Discussion
•  
   Concept Drift
•  
  Foundation
•  
   Definition
•  
   Types of Concept Drift
•  
   Patterns of Change
•  
   Model Evaluation
•  
   Challenges
•  
  Related Work
•  
   Drift Detection
•  
   Sliding Window
•  
   Bagging Ensembles
•  
   State-of-the-art Methods
•  
   Taxonomy
•  
   Quantifying Concept Drift
•  
   Prerequisites
•  
   Test for Real Drift
•  
   Test for Virtual Drift
•  
   Drift Degree
•  
   Datasets
•  
   Experiments
•  
   Discussion
•  
   Self-Adjusting Memory (SAM)
•  
   Architecture
•  
   Time Complexity
•  
   Speedup via Approximate ITTE
•  
   Experiments - SAM-kNN
•  
   Experiments - SAM-NB
•  
   Discussion
•  
   SAM-Ensemble (SAM-E)
•  
   Architecture
•  
   Parallel Implementation
•  
   Datasets
•  
   Experiments
•  
   Discussion
•  
   Real-World Applications
•  
  Interactive Online Learning on a Mobile Robot
•  
   Application Setup
•  
   Experiments
•  
   Discussion
•  
   Personalized Maneuver Prediction
•  
   Dataset
•  
   Experiments
•  
   Discussion
•  
   Personalized Human Action Classification
•  
   Online Action Classification
•  
   Dataset
•  
   Experiments
•  
   Discussion
•  
   Conclusion
•  
   High-Level Insights
•  
   Outlook
•  
  Appendix
•  
  Detailed results
•  
   Motion Classification
•  
   Practice-oriented survey
•  
  Datasets
•  
   Artificial
•  
   Real-world
•  
   Bibliography