This course is about evaluating and selecting features and models. It covers the following topics: evaluation metrics, feature engineering, feature selection, the role of regularization, loss functions, and feature importances.
Course Learning Outcomes
By the end of the course, students are expected to be able to
- build, debug, appropriately evaluate, and refine supervised machine learning models
- reason to some extent the choice of a machine learning model
- explain different feature selection methods and carry out feature selection
- broadly describe and carry out feature engineering
- explain and carry out L1- and L2-regularization