Speaker: Sara Barahona Quirós.

Abstract: Explainable Machine Learning (XAI) refers to the development of machine learning models and algorithms that not only make accurate decisions but also provide understandable and interpretable explanations for those predictions. In traditional machine learning, particularly with deep neural networks and other black box models, understanding how a model arrives at a specific prediction can be challenging. This lack of transparency raises concerns, especially in critical applications such as healthcare, finance, and autonomous systems. Explainable Machine Learning aims to address these concerns by providing human-understandable explanations for the decisions made by machine learning models. This seminar covers key methods for obtaining post-hoc interpretations of black box models, including both global and local approaches. To a better understanding, we present a case study employing the Python libraries LIME and SHAP. Additionally, different methods for interpreting Convolutional Neural Networks are presented, focusing on understanding how these complex models learn.