The paper titled "A probabilistic view on Riemannian machine learning models for SPD matrices" presents a novel approach by integrating various machine learning techniques for Symmetric Positive Definite (SPD) matrices into a probabilistic framework. This integration is achieved through the use of Gaussian distributions defined on the Riemannian manifold, which allows for a reinterpretation of popular classifiers as Bayes Classifiers. By framing these classifiers probabilistically, the study advances the field of Riemannian machine learning, offering new insights and methodologies. The approach highlights the potential of probabilistic models to enhance understanding and performance in handling SPD matrices, which are common in many applications. This work supports the claim that adopting a probabilistic framework represents a significant advancement in Riemannian machine learning. Overall, the paper contributes to the ongoing development of machine learning techniques by providing a fresh perspective grounded in probability theory and differential geometry.