Zurich AI

The training week provides a comprehensive introduction to Explainable Artificial Intelligence (XAI), emphasizing the methodologies and practical applications of cutting-edge models such as PDP, ICE plots, LIME, SHAP, and others. Participants will explore how these techniques enhance interpretability and transparency in AI systems, along with the challenges they face, such as scalability, interpretability trade-offs, and accuracy limitations.
The course also investigates the limitations and reliability of XAI models when applied to complex datasets, with advanced discussions on their performance and practical constraints. A distinctive focus is placed on financial applications, examining how XAI can address the unique challenges and regulatory requirements of the financial sector.
By the end of the course, participants will gain the expertise to implement XAI models, critically evaluate their effectiveness, and apply them responsibly within financial systems, fostering trust and compliance with regulatory standards.

Prior Knowledge: Participants are expected to have a foundational understanding of machine learning concepts, including supervised and unsupervised learning, common algorithms, and evaluation metrics. Certain familiarity with Python programming is needed, with experience using libraries such as scikit-learn, pandas, and NumPy. Familiarity with basic statistics and linear algebra will also be helpful for understanding the mathematical foundations of explainability methods.