The course trees shown below are designed to help you understand the range of Princeton courses offered in Data Sciences, Machine Learning, Optimization, Probability and Statistics, and to assist students and faculty advisors in planning course selections. All of the courses listed are full semester, graded courses.
Please note that courses may not be offered every semester nor every year. In addition, some courses have limited enrollment and fill up quickly. To check if a course is being courses offered in a particular semester, and its enrollment cap, please check the Registrar’s website, or the website of the home department.
If you have any questions or see any errors or omissions, please send us an email.
|SML Graduate Certificate Course Table|
Machine Learning Core Courses (at least one course from the list below)
Statistics Core Courses (at least one course from the list below)
Electives (one from the list below, or an additional course from the core lists above, with permission.)
Data Science and Statistics
NOTES (Graduate Certificate)
- Courses may have distinct prerequisites, and not all courses are offered every year. Please check the registrar’s website.
- COS 524, ECE 535, and COS 511 are sufficiently distinct to be considered complementary courses.
- COS 524 vs ECE 535. Both courses cover ML algorithms, but differ in focus and many (but not all) of the specific topics. COS 524 focuses on the problem of analyzing large complex data sets including text, images, and biological data and has a major data analysis final project. ECE 535 is a proof based course focusing on the theoretical underpinnings and properties of ML algorithms. It has regular theory problem sets, programming assignments (Python), a midterm, and a final exam.
- ECE 436 Machine Learning Theory (UG course tables) covers algorithmic independent bounds on ML. It is complementary to both COS 524 and ECE 535.
- The courses COS/SML 513 and SML/AST 505 are related but differ in several ways. Both courses use Bayesian approaches, emphasize fundamental concepts, and end with a final project in lieu of an exam. COS/SML 513 covers probabilistic model specification and evaluation and efficient computational algorithms for prediction and decision making. SML/AST 505 introduces a statistical approach to data and leverages probabilistic programming frameworks for quickly performing analyses in applications across various scientific disciplines.
- ORF 524 is a broad introduction to modern statistical methods from a theoretical perspective. The course covers statistical theory and methods for point estimation, confidence intervals (including modern bootstrapping), and hypothesis testing. It has homeworks, a midterm, and a final exam. It is complementary to COS/SML 513 and SML/AST 505.
- POL 572 covers applied regression analysis in cross-section settings. It introduces the basic principles of causal inference and then covers various statistical techniques including linear regression and instrumental variables. It offers a more focused perspective on the use of modern statistics for the study of social data.