Automated machine learning for multi-label classification / Marcel Wever ; [Reviewers: Prof. Dr. Eyke Hüllermeier, Prof. Dr. Axel-Cyrille Ngonga Ngomo]. Paderborn, 2022
Inhalt
- Titlepage
- Zusammenfassung
- Abstract
- 1 Introduction
- 2 Preliminaries
- 2.1 Introduction to Automated Machine Learning
- 2.1.1 Machine Learning Pipelines
- 2.1.2 General Structure of AutoML Systems
- 2.1.3 Reduction to Hyper-Parameter Optimization
- 2.1.4 Grammar-Based Search
- 2.1.5 Meta-Learning
- 2.1.6 Neural Architecture Search
- 2.2 Introduction to Multi-Label Classification
- 3 ML-Plan: Automated Machine Learning via Hierarchical Planning
- 4 ML-Plan for Unlimited-Length Machine Learning Pipelines
- 5 Automating Multi-Label Classification Extending ML-Plan
- 6 AutoML for Multi-Label Classification: Overview and Empirical Evaluation
- 7 LiBRe: Label-Wise Selection of Base Learners in Binary Relevance for Multi-Label Classification
- 8 Ensembles of Evolved Nested Dichotomies for Classification
- 9 Predicting Machine Learning Pipeline Runtimes in the Context of Automated Machine Learning
- 10 Conclusion and Open Questions
- 11 Epilog – On-The-Fly Computing for Machine Learning Services
- Further References
- Own Publications
- List of Figures
- Colophon