Practical Machine Learning

3 credits · Graduate

This course provides a practical introduction to machine learning, emphasizing how to build, analyze, and apply models in real-world settings. We begin with regression and model evaluation, then study regularization and cross-validation for controlling overfitting and selecting models. We develop a probabilistic view of learning through maximum likelihood estimation and Bayes estimators, connect classification to logistic regression and cross-entropy, and finally connect these ideas to training neural-network based models. The course also covers gradient-based optimization, dimensionality reduction, and introduction to deep learning workflows in PyTorch. Students gain hands-on experience through coding assignments and projects.