Explore a 48-minute lecture on non-asymptotic analysis of l1 support vector machines presented by Jan Vybiral at the Hausdorff Center for Mathematics. Delve into the importance of support vector machines as a crucial machine learning algorithm for classification problems. Discover how replacing l2 regularization with l1 regularization leads to sparse classifiers for high-dimensional data, which is vital in many real-world applications. Gain insights into the analysis of l1-support vector machines, including error guarantees valid in non-asymptotic regimes. Learn about the collaborative work between Jan Vybiral and Anton Kolleck from TU Berlin. This lecture, part of the Hausdorff Trimester Program on Mathematics of Signal Processing, offers valuable knowledge for those interested in advanced machine learning techniques and their mathematical foundations.