Learning theory is concerned with the analysis of learning algorithms, in particular regarding their intrinsic limitations or their consistency. Another major topic in learning theory is how we can obtain performance guarantees for the algorithms. These guarantees typically take the form of generalization error bounds.