This advanced course focuses on the design and implementation of machine learning methods and model risk management practices for financial and insurance applications. Topics include supervised learning (GLMs, survival models, decision trees, ensemble methods, and neural networks), unsupervised learning (PCA and clustering), and reinforcement learning. Students will also explore modern paradigms such as transformer architectures and large language models, including fine-tuning and prompt engineering.

The course emphasizes model development, refinement, and customization, oftentimes from first principles, across different stages of the data science lifecycle, building upon the data-preparation and workflow foundations established in Data Science and Machine Learning I. This course places a heavier emphasis on programming and applied implementation using modern data science libraries and tools.