Foundations of Model Building

This unit develops practical skills for preparing data, training and evaluating machine learning models, and applying essential tooling and safeguards. Students will practice data structures and notebook workflows, examine algorithm behavior and generalization, and address privacy, statistical, and ethical considerations to prepare for advanced model scaling and deployment.

Learning Objectives

• Apply appropriate data structures (sets, lists, tuples, dictionaries) and Jupyter Notebook features (magic functions, cells, and indexing) to ingest, organize, and preprocess datasets for machine learning pipelines.
• Analyze model training and generalization by computing and interpreting out-of-sample errors, cross-validation results, and sanitization impacts to detect and mitigate overfitting.
• Demonstrate implementation and comparison of core algorithms — including greedy decision trees, ensemble (weak learner) techniques, distance-based clustering, basic neural networks, and feature-type mixing — and explain their behavior and interpretability trade-offs.
• Evaluate statistical assumptions, data privacy concerns (secret management and sanitization), and ethical/future-technology risks to make responsible decisions in model design and application.