Requirements for the Master's degree within the PhD program


Students working towards a PhD must first obtain their Master's Degree within the PhD program. The process leading to this degree is part of the PhD program and must not be confused with the Professional Master's Program offered by this department. To obtain a Master's Degree, students in the PhD program must meet the following requirements:

All students must complete an approved sequence of nine courses. An approved sequence consists of five core courses, and four electives. Students taking the Computational Math track are required to take a different set of core courses and will choose from a different set of electives. Click here for these and other specifics of the Computational Math track.

The Core

The set of core courses has been designed to bring sharp focus on the foundations to the program, guarantee sufficient breadth, and foster collegiality among our graduate students. Each student selects a set of five courses from the list below; the selection must include one course in Machine Learning, two courses in Systems, and two courses in Theory.

Core courses: Machine Learning

  • CMSC 35400 - Machine Learning
  • TTIC 31020 - Introduction to Statistical Machine Learning

Core courses: Systems

  • CMSC 23000 - Operating Systems (permission required)
  • CMSC 23700 - Introduction to Computer Graphics (permission required)
  • CMSC 32200 - Computer Architecture
  • CMSC 32630 - Advanced Implementation of Computer Languages
  • CMSC 33100 - Advanced Operating Systems
  • CMSC 33300 - Networks & Distributed Systems
  • CMSC 33520 - Data Intensive Systems
  • CMSC 33550 - Introduction to Databases
  • CMSC 33710 - Scientific Visualization

Core courses: Theory

In Spring 2018, the list of Theory options has been greatly expanded. The updated list is described below. It includes several undergraduate courses.

Students choose two out of the following list of courses, including one from the first group and one from the second group.

First group

Any of the courses in this group is sufficient as a prerequisite to Algorithms.


  • CMSC 37115** - Introduction to mathematical reasoning via Discrete Mathematics. (Offered each autumn.) Students with little prior experience with rigorous mathematical reasoning are encouraged to take this course in the first quarter of their studies. They will practice the rudiments of mathematical thinking and problem solving through accessible subjects of relevance to most branches of computer science, including basic number theory, counting, discrete probability, graph theory, asymptotic rates of growth. (The course will not include linear algebra.)
  • CMSC 27100 - Discrete Mathematics. -- This undergraduate course is likely to be heavily oversubscribed, so we ask PhD students and advisors to consider alternatives.
  • CMSC 27130 - Honors Discrete Mathematics.
  • CMSC 31150 - Mathematical Toolkit. -- This graduate course, offered by TTIC, focuses on Linear Algebra and Probability Theory. It is especially recommended for students interested in Machine Learning and in Theory.
  • CMSC 38400 - Cryptography. -- This graduate course is co-located with the undergraduate course "Introduction to Cryptography" (CMSC 28400).

Second group


  • CMSC 37000 - Algorithms. -- Offered by TTIC.
  • CMSC 28100 - Introduction to Complexity Theory. Offered in Winter 2019.
  • CMSC 38130** - Complexity Theory. This graduate course will be co-localed with the undergraduate "Honors Complexity Theory." Recommended for students interested in Theory. Offered in Autumn 2018.
  • CMSC 27500 - Graph Theory. -- This course will not be offered in 2018/19 but we hope to be able to offer it thereafter in alternate years.
  • CMSC 37530** - Graph Theory. -- This graduate course will be co-located with the undergraduate "Honors Graph Theory." Recommended for students interested in Theory. Offered in Spring 2019 and in alternate years thereafter.
  • CMSC 37200 - Combinatorics. -- This graduate course will be co-located with the undergraduate "Honors Combinatorics (CMSC 27410)." Recommended for students interested in Theory. Offered in Spring 2020 and in alternate years thereafter.


The list of courses that are available to serve as electives varies significantly year-to-year. The following courses are approved as electives for 2016-2017. To ensure sufficient breadth, students must take electives from either (a) two or more areas or (b) from the computational mathematics list.

Electives: Artificial Intelligence

  • CMSC 45300 - Machine Learning
  • TTIC 31020 - Introduction to Statistical Machine Learning
  • TTIC 31050 - Introduction to Bioinformatics and Computational Biology
  • TTIC 31120 - Statistical and Computational Learning Theory
  • TTIC 31040 - Introduction to Computer Vision
  • TTIC 31170 - Planning, Learning, and Estimation for Robotics and Artificial Intelligence
  • TTIC 31210 - Advanced Natural Language Processing
  • TTIC 31220 - Unsupervised Learning and Large-Scale Data Analysis

Electives: Computational Mathematics

  • CMSC 34900 - Solving PDEs using the FEniCS Project
  • CMSC 30900 - Mathematical Computation I: Matrix Computation

Electives: Systems

  • CMSC 33250 - Introduction to Computer Security
  • CMSC 33301- Topics in Systems  Intermittent Computing  Desktops to the Cloud
  • CMSC 33550 - Introduction to Databases
  • CMSC 33710 - Scientific Visualization
  • CMSC 33520 - Data Intensive Computing Systems
  • Any course from the systems core course list that is not used by the student to fulfill the core requirements.

Electives: Theory

  • Any course from the theory core course list that is not used by the student to fulfill the core requirements.
  • CMSC 38700 - Complexity Theory B
  • CMSC 38500 - Computability and Complexity Theory
  • CMSC 39010 - Computational and Metric Geometry
  • CMSC 39600 - Topics in Theoretical Computer Science
  • CMSC 38815 - Geometric Complexity
  • CMSC 35600 - Algorithms in Finite Groups

Topics courses can be used more than once, provided the material taught in the different offerings is distinct, and the course involves structured and graded work.

Students may petition the Graduate Committee to substitute other courses for those listed. Students are required to submit their petitions for substitution before they take a course with which they intend to fulfill the electives requirement.

Grade Requirements

There are specific grade requirements for both core courses and electives (described below). The spirit of these requirements can be summed up by the following motto: a student must demonstrate proficiency in all areas and excellence in at least one area.

The minimum formal requirements for the core courses ("Ph.D. Pass") are the following: Students are required to complete the five core courses with a grade point average (GPA) of at least 3.25 in the five core courses. In computing the GPA, A=4, B=3, and a + or a - counts as .3 of a point. Note that for the core courses, students who significantly outperform even the typical "A" students may receive a grade of "A+" (recorded internally by the CS Department Student Representative--the University does not officially grant the grade of A+.) So, for instance a student with grades A+, B+, B+, B-, B- in the five core courses has a GPA of 3.26 and thus satisfies the minimum GPA requirement, as does a student with grades of A+, A, B+, B, and C-. In the graduate program grades below C- are not passing grades.

Students must complete their electives with a grade of B or better in each course.

The previous rules, that apply to the classes before the cohort first enrolled in Summer or Autumn 2015 can be found here.

Students who fail to meet the core course requirements stated in the preceding paragraph may continue on to write a master's paper and complete a master's degree, if they meet the following requirement ("Master's Pass"): complete all the five core courses by the end of the spring quarter of the second year with a grade of at least C- in each core course and with a grade point average (GPA) of at least 3.00 in the five core courses. Such students will be supported for at most one quarter of their third year.

Students who do not meet the Ph.D. Pass requirements for these courses cannot continue their studies beyond autumn quarter of their third year. Students who do meet these minimum requirements will not automatically be allowed to continue after their third year; the faculty will decide continuation based on the student's perceived capacity to perform Ph. D. level independent research in a specific area.

Master's Paper and Exam

Each student must complete a Master's paper that must demonstrate knowledge of a particular area of computer science, including in-depth familiarity with the related literature.

Students must give public presentations of their Master's papers, followed by a private exam. At the public presentation and in the private exam, students must be able to give detailed answers to questions about the work described in their paper.