Introduction.- Part I, Basic Architecture of Metalearning and AutoML Systems.- Metalearning Approaches for Algorithm Selection I.- Evaluating Recommendations of Metalearning / AutoML Systems.- Metalearning Approaches for Algorithm Selection II.- Automating Machine Learning (AutoML) and Algorithm Configuration.- Dataset Characteristics (Metafeatures).- Automating the Workflow / Pipeline Design.- Part II, Extending the Architecture of Metalearning and AutoML Systems.- Setting Up Configuration Spaces and Experiments.- Using Metalearning in the Construction of Ensembles.- Algorithm Recommendation for Data Streams.- Transfer of Metamodels Across Tasks.- Automating Data Science.- Automating the Design of Complex Systems.- Repositories of Experimental Results (OpenML).- Learning from Metadata in Repositories.