Carnegie Mellon University

Joint PhD Program in Statistics & Machine Learning

This PhD program differs from the ML PhD program in that it places significantly more emphasis on preparation in statistical theory and methodology. Similarly, this program differs from the Statistics PhD program in its emphasis on machine learning and computer science. (See below for a more details on the course requirements.)

Students in this track will be involved in courses and research from both the Departments of Statistics and Machine Learning. During the first year, students will normally be situated in the Department of Statistics. During later years, students will normally be located in the Machine Learning Department unless the primary advisor is in the Department of Statistics. In years 2 and after thesis research co-supervised by a faculty in ML and a faculty in Statistics, or supervised by a joint faculty member. The thesis committee must contain at least one member with home department of Statistics and one with home department of ML.

The typical curriculum is as follows:
(10- designates a ML course. 15- designates a CS course. 36- designates a statistics course.)

* indicates a course that is in the joint program but not in the ML PhD program.
# indicates a course that is in the joint program but not in the Statistics PhD program.

Generally, these courses replace electives in the ML PhD program. The exception is 36-757/758 which serves as the research course 10-920.

FALL - Year One

SPRING - Year One

#10-715 Advanced Introduction to  Machine Learning #10-716 Advanced Machine Learning: Theory & Methods
  36-705 Intermediate Statistics *36-752 Advanced Probability Overview
*36-707 Regression Analysis *36-757 Advanced Data Analysis (ADA) I

FALL - Year Two

SPRING - Year Two

*36-755 Advanced Statistics One of the following courses:
#10-708 Graphical Models
#10-725 Convex Optimization
#15-826 Multimedia Databases
#15-750 Algorithms or 15-853 Algorithms in the Real World
*36-758 ADA II

*36-750 Statistical Computing (recommended)