In recent years, Machine Learning (ML) has been used in novel applications—from generating new art and music to systems that accurately and reliably predict outcomes of medical interventions in real-time. Faster computing hardware, large datasets, and the probabilistic paradigm of ML, which frames advances like neural networks within statistical learning, have enabled these developments. In this course, we introduce the foundational concepts behind the probabilistic paradigm of predictive ML: statistical model specification and learning. We will focus on connecting theory with real-world applications. Students will get hands-on experience building models for specific tasks, most taken from healthcare contexts, using probabilistic programming languages. While expanding our methodological toolkit, we will simultaneously introduce critical perspectives to examine the ethics of ML within sociotechnical systems. This course lays the foundation for advanced study and research in ML. Topics include: directed graphical models, deep regression/classification, frequentist learning, and model evaluation. For more information, see the course website: https://mogu-lab.github.io/cs245/.

Units: 1

Max Enrollment: 18

Prerequisites: One of the following (CS 230, CS 230P, CS 230X) and MATH 115, and one of the following (MATH 205, MATH 206, MATH 220, MATH 225, STAT 218, or STAT 318), and permission of the instructor.

Distribution Requirements: MM - Mathematical Modeling and Problem Solving

Degree Requirements: DL - Data Literacy (Formerly QRDL)

Typical Periods Offered: Spring

Semesters Offered this Academic Year: Spring

Notes: This course will not be offered in the 27-28 Academic Year.