Learn how to quantify uncertainties in deep learning applications for astroparticle physics through this comprehensive lecture by Christian Glaser. Explore the critical importance of understanding and measuring uncertainty in neural network predictions when applied to high-energy particle detection and cosmic ray analysis. Discover various uncertainty quantification techniques including Bayesian neural networks, Monte Carlo dropout, and ensemble methods specifically tailored for astroparticle physics datasets. Examine real-world case studies demonstrating how proper uncertainty estimation improves the reliability of deep learning models in detecting neutrinos, cosmic rays, and other high-energy particles. Understand the challenges unique to astroparticle physics data, including sparse signals, high noise levels, and the need for robust statistical inference. Master practical implementation strategies for incorporating uncertainty quantification into existing deep learning pipelines used in particle physics experiments and observatories.