Graduate Program Handbook (7/16/2013)

1 Introduction

This handbook is for graduate students in the Master of Science (MS) program and the Doctor of Philosophy (Ph.D.) program in computer science at Georgetown University. It documents the academic regulations for these programs, but it does not replace or supplant the academic regulations documented in the Graduate Bulletin. Students should always read the relevant sections of the Bulletin before consulting this handbook. Students are responsible for knowing and following the academic regulations described in the Bulletin and in this handbook. If students are unsure about regulations or how they apply, then they should discuss any uncertainty with the Director of Graduate Studies (DGS) or with someone in academic affairs in the Graduate School. Students who are unaware of academic regulations are still subject to them.

2 Master’s Program

The MS program, through course-work and thesis options, lets students strengthen their foundational education, prepare for technical careers in industry, or prepare for advanced study at the doctoral level. The department’s faculty work in the areas of algorithms, artificial intelligence, bioinformatics, computer and network security, database systems, data mining, distributed algorithms, distributed systems, information assurance, information retrieval, machine learning, networking, non-standard parallel computing, and parallel algorithms.

2.1  Degree Requirements

Students elect to complete the requirements of the degree by taking ten courses (30 credits) or by taking eight courses (24 credits) and writing a thesis. All students must take Algorithms (COSC-540) and Architecture (COSC-520). Students pursuing the course-work option take a total of eight electives to complete the degree. A generic schedule for a full-time student pursuing the course-work option appears in Table 1.

Students may choose as an elective any course numbered 350 or higher, but at least five of these classes must be numbered 500 or higher. The department regularly offers introductory or advanced electives in the areas of artificial intelligence, computer and network security, database systems, data mining, information assurance, information retrieval, and machine learning.

2.1.1  External Electives

Students can substitute up to two courses from another department for similarly numbered electives, provided that the courses support the student’s plan of study and have been approved by the student’s faculty advisor and the DGS. The department maintains a list of pre-approved external electives. Students may petition for the use of other courses as external electives, but in addition to satisfying the previous criteria, such courses must be approved by the DGS. Students must obtain all necessary approvals before enrolling in the class.

2.1.2  Thesis Option

Students choosing to write a thesis complete similar requirements, but substitute Graduate Thesis Research (COSC-999) for two electives numbered 350–499. That is, thesis students complete the core requirements, take no more than one elective numbered 350–499, and take at least five electives numbered 500 or higher, for a total of twenty-four credit hours. Students selecting the thesis option must have a grade-point average of 3.4 or higher prior to their first enrollment in Thesis Research (COSC-999) and after the completion of their Master’s course work. A hypothetical schedule for a full-time student pursuing the thesis option appears in Table 2.

2.2  Tutorial Courses

Tutorial courses are opportunities for MS students to pursue independent study under the direction of a professor on a topic of their choosing. Tutorials are for the further development of a research project begun as part of a traditional graduate lecture that may lead to a thesis proposal. Master’s students are limited to one tutorial course.

2.3  Registration in Thesis Research

Thesis Research is a registration category that qualifies thesis students for full-time enrollment after they have completed their course work and while they are working on their thesis project. Work on one’s thesis can begin at any time, but registration in Thesis Research (COSC-999) occurs only after students have completed all courses or during the semester in which they complete all courses. Students should enroll in Section 1 of Thesis Research (COSC-999-01) if they have completed all courses. They should enroll in Section 3 of Thesis Research (COSC-999-03) if they are taking their final courses and these courses alone do not constitute full-time enrollment. After their initial registration in Thesis Research, subsequently students are automatically enrolled in Section 1 until they graduate.

2.4  Thesis Advisor

Full-time students should identify a topic area (or topic) and a thesis advisor before the midpoint of their second semester. Part-time students should do so before the midpoint of their fourth semester. The thesis advisor must be a full-time faculty member in the Department of Computer Science and research-active in their area of computer science as demonstrated by top-tier conference and journal publications. The thesis advisor need not be the same person advising the student on curricular issues.

2.5  Thesis Committee

After identifying a thesis advisor and topic, students work with their thesis advisor to form a committee consisting of at least two additional members who are full-time faculty in the Department of Computer Science, research-active in their area of computer science, and qualified to supervise, guide, review, and judge critical aspects of the thesis. If appropriate, one member of the review committee can be such a person from another department at Georgetown or another university, or from industry or government. The DGS approves the committee by signing the Thesis Proposal Form.

2.6  Thesis Proposal

Research for the thesis and its proposal should begin during the first academic year. Full-time students must submit their thesis proposal before the end of their second semester. Part-time students must do so before the end of their fourth semester. Students must complete and submit a Thesis Proposal Form to the Graduate School after obtaining all necessary approvals of the advisor, committee, and DGS.

2.7  Thesis Review

Once students complete a final draft of their thesis, with their advisor’s approval, they distribute the thesis to the other members of their committee for a period no shorter than three weeks. After the members of the committee complete their review, they assess the thesis using the department’s rubric. The advisor distributes copies of the assessments to the student, the members of the committee, and the DGS, and oversees any necessary changes to the thesis. Once the changes to the thesis have been completed, students complete the Graduate School’s Thesis Reviewer’s Report and obtain all necessary approvals of the advisor and committee. Students then submit the completed form with a draft of their thesis to the DGS.

After the review period, students work with their advisor, their committee, the department’s Colloquium Chair, and the DGS to schedule a public presentation of the work, the details of which must be finalized and agreed upon two weeks in advance of the defense date. After all parties agree with the schedule, the DGS submits the completed Thesis Reviewer’s Report to the Graduate School at least one week prior to the date of the defense. See Section 4.13 on the timing of reviews and defenses.

2.8  Thesis Defense

For the public defense of the thesis, students should plan a 45-minute presentation with 15 minutes reserved for public questions. The DGS will supply the Thesis Defense Report Form. Students should bring to the public defense the Graduate School’s Master’s Thesis Cover Sheet. The student, advisor, the members of the committee, and the DGS or a designee must be physically present for the public defense of the thesis.

The members of the committee and the members of the faculty in the audience assess the presentation using the department’s rubric. After the period of public questions, the advisor excuses members of the audience so the committee can privately question the student. The advisor then excuses the student so the committee can deliberate in private about the sufficiency of the thesis and any required modifications. The student must pass with a majority vote.

After these deliberations, the advisor invites the student to rejoin the discussion and delivers the committee’s decision as well as any required modifications to the thesis. If the thesis is acceptable, then the advisor, committee, and student agree to any required modifications, which the advisor should communicate to the committee and the student via email after the conclusion of the defense. The advisor and committee complete and sign the Master’s Thesis Cover Sheet and Thesis Defense Report Form, and return both to the DGS. The advisor distributes copies of the assessments to the student, the members of the committee, and the DGS. The DGS signs the Thesis Defense Report Form and delivers it to the Registrar.

The members of the faculty in the audience who assess the presentation give their completed rubrics to the DGS. The DGS summarizes the scores and sends them along with any comments to the student, the advisor, and the committee.

The thesis advisor oversees any required modifications to the thesis. Once the student completes these modifications to the satisfaction of the advisor and committee, the thesis advisor should notify the DGS so the DGS can sign the Thesis Cover Sheet and deliver it to the Graduate School. At this point, the student can begin the final formatting of the thesis.

2.9  Final Formatting of the Thesis

Once students pass their defense and complete all required modifications, they then put the thesis in the Graduate School’s required format. For more information on this process, see the Graduate School’s page on theses and dissertations. For formatting theses, the Graduate School provides a Word template and a LaTeX style sheet.

2.10  Minimum Requirements

All Master’s students must maintain and complete their course work with a grade-point average of at least 3.0. Thesis students must maintain and complete their course work with a grade-point average of at least 3.4. Students supported through merit-based scholarships must maintain a grade-point average of at least 3.78.

Full-time Master’s students pursuing the course-work option should be enrolled in nine or more credits every semester, except for the last semester in which they complete their degree requirements. Full-time Master’s students pursuing the thesis option should be enrolled in nine or more credits or in the appropriate section of thesis research every semester until completing their degree requirements. Part-time Master’s students must be enrolled in at least three credits or in the appropriate section of thesis research every semester until completing their degree requirements.

2.11  Milestones for Thesis Students

The department expects full-time Master’s students who elect to write a thesis to complete their degree requirements in two years and expect part-time students to do so in three years. To ensure that students complete their degree requirements within these time limits, the department holds students to the following milestones.

Full-time MS students who elect to write a thesis must submit a thesis proposal before the end of their second semester. They must finish all of their courses before the end of their third semester. They must submit the final draft of their thesis to their committee for review by midpoint of their fourth semester. The midpoint of the fall semester is October 15, and the midpoint of the spring semester is March 15. Finally, students must defend their thesis before the Graduate School’s deadline for that semester.

Part-time MS students who elect to write a thesis must submit a thesis proposal before the end of their fourth semester. They must finish all of their courses before the end of their fourth semester. They must submit the final draft of their thesis to their committee for review before the start of classes of their sixth semester. Finally, students must defend their thesis before the Graduate School’s deadline for that semester.

The Graduate Committee reviews each student’s progress at each milestone. Students who fail to reach milestones on time or are not making sufficient progress will be switched to the course-work track so they can complete the remaining courses and graduate on time. Students who miss any milestone must petition the Graduate Committee to obtain permission to continue work on their thesis.

2.12  Time Limits

Full-time Master’s students are expected to complete all requirements for the degree in two years. Part-time students are expected to do so in three years. International students, who must be full-time students, normally have two years to complete the requirements for the degree.

2.13  Extensions of Time

Master’s students must petition to go beyond a third academic year, and there must be complicating circumstances. Students must submit a statement describing the reasons for the request and documentation of the complicating circumstances. Thesis students must also submit a description of the current state of the project, letters of support from the members of the committee, a time line of remaining milestones, and the current draft of the thesis. Students must submit their petition to the DGS before the midpoint of their final semester. Such extensions of time are designed to accommodate short interruptions in progress of duration less than one semester, and are not designed to supplant personal, military, or medical leaves of absence. Students may request extensions of time through their advisor, the DGS, and the Graduate School using the Student Petition for Change to Graduate Program. See the Graduate Bulletin for more information about requests for extension of time limits and leaves of absence.

2.14  MS Students Applying to the PHD program

Students enrolled in Georgetown’s MS program in computer science who would like to change to the Ph.D. program must apply for admission by following the normal application process.

3 Ph.D. Program

The doctoral program in computer science prepares students for research and teaching careers in academia and for research and technical careers in industry and government. The primary areas of concentration of the program are:

• computer and network security
• information search and retrieval
• machine learning and data mining
• networking and distributed systems
• parallel and distributed algorithms

The department’s faculty work in the areas of algorithms, artificial intelligence, bioinformatics, computer and network security, database systems, data mining, distributed algorithms, distributed systems, information assurance, information retrieval, machine learning, networking, non-standard parallel computing, and parallel algorithms.

3.1  Degree Requirements

There are six main requirements for the Doctor of Philosophy program:

1. complete the requirements for the Master’s degree
2. pass three qualifying examinations
3. take three doctoral seminars
4. complete the Apprenticeship in Teaching Program
5. complete and defend a proposal
6. complete and defend a dissertation

3.2  Requirements for the Master’s Degree

Doctoral students must first complete the requirements for the Master’s program, which involves taking ten courses (30 credits) or taking eight courses (24 credits) and writing a thesis. Instead of taking two core courses, doctoral students must take three: Algorithms (COSC-540), Architecture (COSC-520), and Theory (COSC-545). The rules for graduate electives, external electives, and the thesis option apply without modification to doctoral students.

3.2.1  Tutorial Courses

Doctoral students are not eligible for tutorial courses.

3.2.2  Conferral of the MS Degree

Doctoral students who complete the requirements for the MS program receive the degree Master of Science in Computer Science.

3.3  Registration in Thesis Research

Thesis Research is a registration category that qualifies doctoral students for full-time enrollment after they have completed their Master-level course work and while they are working toward the completion of their dissertation. Work on one’s dissertation can begin at any time, but registration in Thesis Research (COSC-999) occurs only after students have completed all courses or during the semester in which they complete all courses for the MS requirement. Students should enroll in Section 1 of Thesis Research (COSC-999-01) if they have completed all courses. They should enroll in Section 3 of Thesis Research (COSC-999-03) if they are taking their final courses and these courses alone do not constitute full-time enrollment. After their initial registration in Thesis Research, subsequently students are automatically enrolled in Section 1 until they graduate.

3.4  Qualifying Exams

Three qualifying exams assess students’ understanding of their Master-level course work, their area of specialty, and their ability to understand and critique scholarly articles. The written qualifying exams consist of one core exam and two area exams.

3.4.1  Core Exam

The core exam is designed to assess a student’s understanding of foundational concepts in the fundamental areas of algorithms, architecture, and theory. Students who have completed all of their core courses are eligible to take the core exam.

The department’s Graduate Committee works with faculty teaching the core curriculum to design, maintain, administer, and grade the core exam. The core exam is based on topic lists for each of the three areas.

Algorithms

The material for the algorithms section of the core exam is based on the course Algorithms (COSC-540), its lecture notes, and its textbook:

• Cormen, T. H., Leiserson, C. E., Rivest, R. L., and Stein, C. (2009). Introduction to algorithms, 3rd edition. The MIT Press.

Math: Students will be expected to know the mathematical concepts used throughout COSC-540, including (but not limited to): order notation, solving recurrences, some easier summations (e.g., geometric and harmonic), and basic probability theory.

Clarification on difficulty (for probability): Students will be expected to know basic formulas including (but not limited to) the definition of expectation, linearity of expectation, and Markov’s inequality, as well as how to use these. Students will not be expected to memorize more complicated formulas like Chernoff bounds, but students should be comfortable using them if they are given.

Algorithms and Data Structures Toolbox: Students will be expected to be familiar with specific solutions for certain problems. It is unlikely there will be (m)any questions that ask students to write down one of these algorithms from scratch. However, students should have an understanding of the solutions as well as their running times. When more than one solution is available, students should also be comfortable choosing which solution is best for a certain setting (which relies heavily on knowledge of running times).

The algorithms students are expected to know include the following. Most of these are algorithms that are typically taught in an undergraduate class. With the exception of a few (notably, the number sorting), these were also reviewed in COSC-540:

• sorting:
  • comparison-based sorting: mergesort, quicksort, and insertion sort.
  • sorting numbers: counting sort, radix sort, and bucket sort.
• graph algorithms:
  • searching: breadth-first search and depth-first search,
  • single-source shortest paths: Dijkstra’s and Bellman-Ford.
  • all-pairs shortest paths: Floyd-Warshall
  • minimum spanning tree: Kruskal’s, Prim’s, and Boruvka’s
  • max flow: Ford-Fulkerson and Edmonds-Karp.
• data structures:
  • balanced (binary) search trees
  • B-trees
  • priority queues and heaps
  • hash tables

Techniques: Students should be comfortable applying the following design and analysis techniques: divide and conquer, dynamic programming, greedy algorithms, randomization, amortized analysis, linear programming and integer-linear programming, approximation-algorithm techniques and analysis, competitive analysis, and techniques for the I/O & cache-oblivious models.

Students should also be comfortable with reducing one problem to another. Some likely reductions are to shortest paths, max flow (e.g., with bipartite matching), and linear programming.

Architecture

The material for the architecture section of the core exam is based on the course Computer Hardware and Systems Architecture (COSC-520), its lecture notes, and its textbooks:

• Patterson, D. A. and Hennessy, J. L. (2011). Computer organization and design: The hardware/software interface, 4th edition, revised. Morgan Kaufmann.
• Hennessy, J. L. and Patterson, D. A. (2011). Computer architecture: A quantitative approach, 5th edition. Morgan Kaufmann.

For the exam, students are expected to know the following topics.

• Basic computer design (see Patterson and Hennessy, 2011, Chapter 4, Appendix C)
  • devices, logic design, and finite-state machines
  • simple multi-cycle processor components and organization
  • simple instruction set architecture
  • language layers and translation
  • OS layering, hardware/software interfaces, entering and exiting the kernel
  • I/O, interrupts, interconnects and busses, DMA, bandwidths and capacities
• Technology trends and their effects on computer architecture
  • device technology trends and Moore’s Law
  • fixed costs, customization, configurability, and profitability
  • hierarchical design, abstraction, and standardized interfaces
  • sales volume, yield, uniformity, and generality
  • Amdahl’s Law and the common case
  • cost versus performance/return
  • energy, power, and infrastructure as systemic limitations
  • speedup and the three parallelism principles: pipelining, interleaving, and multiplicity
• Performance evaluation
  • data dependency graphs, data flow graphs, and tagged dataflow execution
  • the processor performance equation
  • program traces, instruction mixes, benchmark suites
  • performance summaries and comparing means
• Caches
  • bandwidth and latency trends in SRAM, DRAM, and processors
  • memory organization, hierarchies, and power consumption
  • general access maps and associative lookup
  • the cache performance equation
  • cache organization alternatives, pipelined and overlapped operations
  • cache element access and critical paths
  • write and replacement methods and penalty and bandwidth implications
  • program access patterns and performance effects
  • miss rates and block and cache sizes, associativity, and multi-level caches
  • split caches, prefetching
• Virtual Memory (VM)
  • page size, fragmentation, access penalty, miss rates, and program performance
  • address mapping, sharing, and protection
  • I/O buffers and cache coherence
  • translation caching (TLB) and miss rates
  • page table sizes, multi-level and inverted tables
  • data and instruction cache interaction
  • page replacement policies
  • virtual versus physical caches
• Instruction-level Parallelism
  • simple processor pipelining, data and control hazards, pipelined decode and control
  • tracking data dependencies, forwarding, stalling, and nullifying
  • branch prediction and speculation
  • out-of-order execution, multiple issue, concurrent pipelined units
  • loop unrolling, software pipelining, dynamic scheduling
  • instruction issue, reservation stations, instruction commit, data tagging and matching
  • coarse-grained, fine-grained, and simultaneous multi-threading
  • exception handling, state, precise exceptions, pipeline exception control
• Thread-level Parallelism
  • shared memory architectures, private data, asymmetric access
  • cache coherence, snooping and directories
  • relaxed and sequential consistency, program semantics
• Data Parallelism
  • VLIW, EPIC and vectorized ALUs, predicated execution
  • vector operations, multiple lanes, memory banking
  • SIMD versus MIMD versus SIMT, thread scheduling
  • vector processor versus GPU organization

Theory

The material for the theory section of the core exam is based on the course Theory of Computation (COSC-545), its lecture notes, and its textbook:

• Sipser, M. (2012). Introduction to the theory of computation, 3rd edition. Course Technology.

For the exam, students are expected to know the following.

• Students should be comfortable proving that languages are and are not regular and/or context-free. This includes familiarity with all equivalent models for these language classes and the regular language pumping lemma.
• Students should be comfortable with the definitions of different Turing Machine models and the Turing decidable and Turing recognizable language classes. In addition to proving that languages are decidable and/or recognizable, students might also be asked to prove problems not decidable and/or not recognizable using standard reductions and mapping reductions. Students’ reduction tool kits should include the computation history method.
• Students should be comfortable with analyzing time complexity and working with the complexity classes TIME, NTIME, P, and NP. Students should be comfortable proving problems are NP-complete. In addition, students should be comfortable analyzing space complexity and working with the complexity classes SPACE, NSPACE, PSPACE, L, NL, and co-NL. Finally, students should know the main space and time hierarchy theorems.

3.4.2  Area Exams

The area exams are designed to assess a student’s understanding of their area of research and one allied area. There is one exam in each of the program’s areas of concentration, such as information retrieval, parallel and distributed algorithms, machine learning and data mining, and computer and network security. Students who have completed all of their elective courses are eligible to take the area exams.

The DGS appoints a committee of faculty working in each area to design, maintain, administer, and grade the area exams. The committee develops and maintains a reading list, which they review and update each year. The committee designs questions for the area exam based on the reading list. The reading list may identify standard textbooks in the area, or chapters of those books, but the majority of the list’s items are seminal articles in the area. The area exams are based on the reading list and require doctoral students to summarize, evaluate, compare, and criticize the material on the list.

3.4.3  Logistics and Grading

The department offers qualifying exams twice per year in late May and in late August. Students who plan to take one or more exams must report their intent in writing to the DGS before the start of February and June, respectively.

The area exams and the sections of the core exam are two hours in length and are given on separate days of a given week. Students can take some or all of their exams during this week. Exams are proctored by the faculty, who determine the format and rules for the exam.

The authors of the qualifying exams grade them. If the faculty members grading the exams require clarification about any of the answers, then they can interview the student and take the responses into account when making a final decision about the answers. After the exams have been graded, the DGS notifies the students about whether they passed or failed each exam. Students can review the marked exams, but the department keeps and maintains control of all graded exams.

Each of the three sections of the core exam is graded as high pass, worth three points, low pass, worth two points, and fail, worth zero points. Students must earn six or more points on the core exam to pass. The area exams are graded as either passing or failing.

Students must pass all exams before the end of their third year and before they can schedule the presentation and defense of their proposal. Students have two attempts to pass each exam. Students who fail one or more exams must retake them at the next scheduled exam. Students may not switch an examination area after a failure. Students who fail the core exam can retake only the sections they failed, but they must score at least a low pass on each section they retake. Students who fail any exam after two attempts will be dismissed from the program. Students who do not take a scheduled exam fail the exam.

Students elect to take an exam months in advance. They should not elect to take an exam until they are prepared to study for and complete the exam. Once they have elected to take an exam, they must manage their other duties so they can prepare adequantely. For legitimate reasons, with their advisor’s support, students may petition the DGS for a postponement. Legitimate reasons include but are not limited to family and medical emergencies and professional obligations, such as conference travel to deliver a paper. The DGS may grant a first postponement, but subsequent postponements require approval of the Graduate Committee.

Students must take exams on the scheduled date. For legitimate reasons, the proctor may need to reschedule the exam for another day, which must be within one week before or after the scheduled date. For legitimate reasons, the student can ask the proctor to reschedule the exam. The proctor must inform the DGS of the date of the rescheduled exam.

3.5  Doctoral Seminars

The purpose of the doctoral seminars is to expose students to the problems and methods in the most recent literature of different areas of computer science. Doctoral students must take three two-credit, pass/fail doctoral seminars in three different topic areas. Students should take doctoral seminars only after they have completed Master-level course work in the area. Students must take all of the required doctoral seminars before the defense of their dissertation proposal. Doctoral seminars are numbered in the range 800–899.

3.6  Dissertation Advisor

Students should identify a topic area (or topic) and a dissertation advisor no later than their third semester. The dissertation advisor must be a full-time faculty member in the Department of Computer Science and research-active in their area of computer science as demonstrated by top-tier conference and journal publications. The dissertation advisor need not be the same person advising the student on curricular issues.

3.7  Dissertation Committee

After identifying a dissertation advisor and topic, students work with their dissertation advisor to form a committee consisting of at least three additional members who are full-time faculty in the Department of Computer Science, research-active in their area of computer science, and qualified to supervise, guide, review, and judge critical aspects of the dissertation. If appropriate, one member of the committee can be such a person from another department at Georgetown or another university, or from industry or government. The DGS approves the committee by signing the Dissertation Proposal Form.

3.8  Dissertation Proposal

Doctoral students must write a dissertation proposal, defend (or present) the proposal in a seminar open to the public, and obtain their committee’s unanimous approval. The proposal is a serious scholarly work of publishable quality that identifies the problem of interest, related work, how existing approaches are inadequate, the proposed approach, the plan of study, how progress and success is to be measured, and preliminary results. The proposal should be supported with the student’s high-quality, peer-reviewed conference or journal publications.

Once students complete a final draft of their proposal, with their advisor’s approval, they distribute the proposal to the other members of their committee for a period no shorter than three weeks. After the members of the committee complete their review, they assess the proposal using the department’s rubric. The advisor distributes copies of the assessments to the student, the members of the committee, and the DGS. Once the committee determines that the proposal requires only minor modifications, students work with their committee to schedule a defense of the proposal. During this period, the advisor oversees any necessary changes to the proposal. See Section 4.13 on the timing of reviews and defenses.

3.9  Proposal Defense

For the public defense of the proposal, students should plan a 20-minute presentation with 10 minutes reserved for public questions. The student, advisor, the members of the committee, and the DGS or a designee must be physically present for the public defense of the proposal.

The members of the committee and the members of the faculty in the audience assess the presentation using the department’s rubric. After the period of public questions, the advisor excuses members of the audience so the committee can privately question the student. The advisor then excuses the student so the committee can deliberate in private about the sufficiency of the proposal and any required modifications. The student must pass with a unanimous vote.

After these deliberations, the advisor invites the student to rejoin the discussion and delivers the committee’s decision as well as any required modifications to the proposal. If the proposal is acceptable, then the advisor, committee, and student agree to any required modifications, which the advisor should communicate to the committee and the student via email after the conclusion of the defense. The dissertation advisor distributes copies of the assessments to the student, the members of the committee, and the DGS, and oversees any required modifications to the proposal.

The members of the faculty in the audience who assess the presentation give their completed rubrics to the DGS. The DGS summarizes the scores and sends them along with any comments to the student, the advisor, and the committee.

After passing the proposal defense, the student completes the Graduate School’s Dissertation Proposal Form and obtains all necessary approvals of the advisor, committee, and DGS. The DGS must also receive a copy of the final version of the dissertation proposal. Students must make sure that they also satisfy the Graduate School’s requirements for the dissertation proposal.

3.10  Candidacy

Once students have completed the MS requirements, passed three doctoral seminars in different areas, passed the written qualifying exam, and gained their committee’s approval of their proposal, they are admitted to candidacy and are eligible to pursue their dissertation project and participate in the Apprenticeship Teaching program.

3.11  Apprenticeship in Teaching Program

All doctoral students must complete the

1. Introduction to Teaching Resources
2. Assessment and Grading
3. Syllabus Design
4. Effective Classroom Interaction
5. The Teaching Portfolio

To complete the workshop requirements, students participate in two elective workshops that vary from semester to semester. Examples include:

• Learning Styles
• Online Writing
• What does ‘Diversity in the Classroom’ Mean?

Students complete the requirements for the program by completing authentic teaching tasks, which include classroom observation, syllabus design, and a video-taped teaching practice.

When appropriate, doctoral students may teach in the department. Doctoral students may serve as teaching assistants with duties that include holding office hours and grading assignments. They may lead one-credit discussion or laboratory sections in conjunction with three-credit courses that faculty offer. Finally, doctoral students who plan academic careers may be able to teach or co-teach an elective topics course at the undergraduate level toward the end of their program.

3.12  Dissertation Review

Once students complete a final draft of their dissertation, with their advisor’s approval, they distribute the dissertation to the other members of their committee for a period no shorter than three weeks. After the members of the committee complete their review, they assess the dissertation using the department’s rubric. The advisor distributes copies of the assessments to the student, the members of the committee, and the DGS. Once the committee determines that the dissertation requires only minor modifications, students work with their committee to schedule a pre-defense of the dissertation. During this period, the advisor oversees any necessary changes to the dissertation. See Section 4.13 on the timing of reviews and defenses.

3.13  Dissertation Pre-Defense

In the pre-defense, students formally present their dissertation to their committee in private. They must also respond to comments and concerns that arise during the review of the dissertation. Students should plan a 45-minute presentation with 15 minutes reserved for questions. Students should also bring to the pre-defense the Graduate School’s Doctoral Dissertation Reviewers’ Report.

After the presentation, the committee meets privately and votes whether to accept the dissertation. The vote must be unanimous for acceptance. The committee records the vote on the Doctoral Dissertation Reviewers’ Report.

After these deliberations, the advisor invites the student to rejoin the discussion and delivers the committee’s decision as well as any required modifications to the dissertation. If the dissertation is acceptable, then the advisor, committee, and student agree to any required modifications, which the advisor should communicate to the committee and the student via email after the conclusion of the pre-defense. The advisor is responsible for ensuring that the student completes the required changes. The committee also communicates any required modifications to the presentation. Once the changes to the dissertation have been completed, students complete the Graduate School’s Doctoral Dissertation Reviewers’ Report and obtain all necessary approvals of the advisor and committee. Students then submit the completed form with a draft of their dissertation to the DGS.

After the review period, students work with the advisor, committee, the department’s Colloquium Chair, and the DGS to schedule a public defense of the dissertation, the details of which must be finalized and agreed upon two weeks in advance of the defense date. After all parties agree with the schedule, the DGS submits the completed Doctoral Dissertation Reviewer’s Report to the Graduate School at least one week prior to the date of the defense.

3.14  Dissertation Defense

For the public defense of the dissertation, students should plan a 45-minute presentation with 15 minutes reserved for public questions. The DGS will supply the Dissertation Defense Report Form. Students should bring to the public defense the Graduate School’s Doctoral Dissertation Cover Sheet. The student, advisor, the members of the committee, and the DGS or a designee must be physically present for the public defense of the dissertation.

The members of the committee and the members of the faculty in the audience assess the presentation using the department’s rubric. After the period of public questions, the advisor excuses members of the audience so the committee can privately question the student. The advisor then excuses the student so the committee can deliberate in private about the sufficiency of the dissertation and any required modifications. The student must pass with a unanimous vote.

After these deliberations, the advisor invites the student to rejoin the discussion and delivers the committee’s decision as well as any required modifications to the dissertation. If the dissertation is acceptable, then the advisor, committee, and student agree to any required modifications, which the advisor should communicate to the committee and the student via email after the conclusion of the defense. The advisor and committee complete and sign the Doctoral Dissertation Cover Sheet and Dissertation Defense Report Form, and return both to the DGS. The advisor distributes copies of the assessments to the student, the members of the committee, and the DGS. The DGS signs the Dissertation Defense Report Form and delivers it to the Registrar.

The members of the faculty in the audience who assess the presentation give their completed rubrics to the DGS. The DGS summarizes the scores and sends them along with any comments to the student, the advisor, and the committee.

The dissertation advisor oversees any required modifications to the dissertation. Once the student completes these modifications to the satisfaction of the advisor and committee, the dissertation advisor should notify the DGS so the DGS can sign the Doctoral Dissertation Cover Sheet and deliver it to the Graduate School. At this point, the student can begin the final formatting of the dissertation.

3.15  Final Formatting of the Dissertation

Once students pass their defense and complete all required modifications, they then set the dissertation in the Graduate School’s required format. For more information on this process, see the Graduate School’s page on theses and dissertations. For formatting dissertations, the Graduate School provides a Word template and a LaTeX style sheet.

3.16  Minimum Requirements

Doctoral students must maintain and complete their course work with a grade-point average of 3.5. Doctoral students should not earn a grade lower than B-.

3.17  Milestones for Doctoral Students

The department expects doctoral students to complete their degree requirements in no more than five years. To ensure that students complete their degree requirements within these time limits, the department holds students to the following milestones.

Doctoral students must finish their MS-level courses before the end of their second year. They must pass all of their qualifying examinations before the end of the third year. They must successfully defend their proposal before the end of the fourth year. They must ascend to candidacy before the end of their fourth year, which includes completing their MS-level courses, doctoral seminars, qualifying examinations, and proposal. They must submit the final draft of their dissertation to their committee for review within the first thirty days of the semester in which they plan to graduate. Finally, students must defend their dissertation before the Graduate School’s deadline for that semester.

3.18  Time Limits

Five years is the expected time for doctoral students to complete all requirements for the degree. Doctoral students must complete all of the requirements for the degree in no more then seven years. They must complete all Master’s requirements in no more than two years. They must pass all of the qualifying exams before the end of the third year. After ascending to candidacy, doctoral students must complete all remaining degree requirements in no more than five years.

3.19  Annual Reports

At the end of each academic year, doctoral students must submit a one-page report to their advisor. The report should describe the progress made toward the completion of their degree requirements and list significant accomplishments for the period, such as completing one’s course work, passing a qualifying exam, or submitting a paper for publication. The report can also document any professional or personal issues that may have impeded progress. The advisor reviews the report, comments, signs it, and forwards it to the DGS for further review by the Graduate Committee. Students who have not yet identified an advisor submit reports directly to the DGS.

3.20  Regular Meetings with Doctoral Committee

After ascending to candidacy, doctoral students must meet with their advisor and doctoral committee once per semester to discuss their progress on the dissertation project. Written summaries of these meetings should be included in the advisor’s and students’ reports to the Graduate Committee.

3.21  Extensions of Time

Doctoral students must petition to go beyond a fifth academic year. Students must submit a statement describing the reasons for the request, a description of the current state of the dissertation project, a letter of support from the advisor, a time line of remaining milestones, and the current draft of the dissertation. Students must submit their petition to the DGS before the midpoint of their final semester. Such extensions of time are designed to accommodate short interruptions in progress of duration less than one semester, and are not designed to supplant personal, military, or medical leaves of absence. Students may request extensions of time through their advisor, the DGS, and the Graduate School using the Student Petition for Change to Graduate Program. See the Graduate Bulletin for more information about requests for extension of time limits and leaves of absence.

A second extension may be granted, but only in the case of complicating circumstances. In addition to the previous requirements, students must submit documentation of the complicating circumstances and letters of support from the members of the committee. The Graduate Committee must approve the petition with a majority vote.

4 General Guidelines and Regulations

There are a number of issues that pertain to both the Master’s and Ph.D. programs. Among other topics, the following sections deal with faculty mentoring, research ethics, financial aid, and courses taken at institutions other than Georgetown.

4.1  Faculty Mentoring

Faculty mentoring is a critical aspect of graduate education. Advice for graduate students generally divides into four categories: academic regulations, curricular issues, research, and post-graduation planning. As students progress toward the completion of their degree, the type of advising they need and the faculty from whom they can consult for advising changes.

After students decide to matriculate and at the beginning of their program, they can seek advice about all matters from the DGS or the Graduate Coordinator. As students progress in their course work, they can seek advice from their instructors about curricular matters, especially if the instructors work in the area in which they are interested. Professors working in a student’s area of interest are in the best position to provide advice on curricular matters, such as course selection and plans of study.

As students progress from the conception to the completion of a thesis or dissertation, they can consult prospective advisors or their official advisor for guidance and mentoring on all matters. After students form committees, they can seek guidance from its members on relevant aspects of their research project. Crucially, students should recruit members for their committee who can provide such advice. Although students can seek advice from faculty members because of their roles as DGS, instructor, committee member, and advisor, students are free to find and consult mentors and advisors who are not acting in these official roles.

4.2  Scholarly Practice and Academic Integrity

At the most fundamental level, scholarly practice involves giving credit where credit is due. It is the way scholars acknowledge and document the ideas, statements, analysis, and the like, that influenced their own ideas, statements, and analysis. The mechanism for acknowledging and documenting the sources of such information is bibliographic citation.

The faculty assume that graduate students have had basic training in scholarly practice during secondary and post-secondary education. Through course work and research projects, the faculty will help graduate students develop further as scholars and researchers in computer science. However, graduate students are responsible for their own education on proper scholarly practice, especially if they have not yet developed sufficient skills and knowledge.

The faculty have adopted a default honor policy that defines acceptable practice for all of its courses. Professors may modify the default policy to suit their purposes. Professors often make clear with their syllabi and assignments their expectations for documentation, but if they are not clear or if they are not clear for a specific situation, then students are responsible for consulting the professor for clarification.

Graduate students, as scholars in training, are held to a high standard. Failure to follow proper scholarly practice, whether intentionally or unintentionally, is a serious matter. Other violations of academic integrity, such as cheating or fabrication of data, are also serious. The Graduate School urges anyone with a “reason to believe that a graduate student has engaged in academic misconduct…to report…such information…to the Dean of the Graduate School.” See the Graduate Bulletin for more information about the policies and procedures governing academic integrity and academic misconduct.

4.3  Computer Systems Acceptable Use Policy

Graduate students in computer science must pay particular attention to the university’s Computer Systems Acceptable Use Policy. Research in computer science may involve activities that, without thoughtful design or proper approval, may constitute an unacceptable use of the university’s computer systems. Regardless of the objective and importance of a class or research project, students must always use the university’s systems acceptably. Students who conceive of a project that may or will violate the university’s policy on the acceptable use of computer systems must consult with their instructor or advisor about how to design or gain approval for the project so it does not violate the acceptable-use policy. In many cases, professors who teach classes and conduct research that involve issues of acceptable use of computer systems have already obtained the approvals and engineered the networks and systems required for compliance with the university’s policy. Students with questions about whether activities constitute acceptable use should consult with their professor or advisor.

4.4  Research Involving Human Subjects

“Federal law requires that all proposed research involving human subjects first be reviewed by an authorized institutional body in order to ensure that adequate protections are provided to those persons who are participants in or subjects of the proposed research” (Graduate Bulletin, Section IV.B.2). In computer science, research that involves human subjects includes but is not limited to user surveys, usability studies, expert labeling of data sets, and derivation of data sets from Web sites (e.g., social-networking sites). The issue of research on human subjects is likely to arise during a thesis or dissertation project, but it is also an issue for class projects, funded research projects, and independent projects. Regardless of the context, students must apply for and receive approval to conduct all proposed research projects that involve human subjects prior to their start.

Georgetown’s Institutional Review Board (IRB) requires that all students and their advisors conducting research involving human subjects first complete a training course on protecting human subjects. After completing the course, students must file an application for approval to the appropriate IRB. For more information, students should consult their advisors and Georgetown’s IRB.

The Graduate Bulletin requires students to include an approved IRB application with their proposal if the project involves human subjects. If students learn after the submission of the proposal that their project requires the involvement of human subjects, then they must apply for and obtain approval for the study through the IRB before proceeding further with the project. If the IRB approves the application, then students must submit the approved application and the IRB’s letter granting approval to the DGS and to the Graduate School. If the IRB does not approve the application, then students must pursue another project or redesign the project and reapply for approval.

4.5  Financial Aid

The department provides two forms of financial aid to graduate students during the academic year: merit-based scholarships and research assistantships. Merit-based scholarships cover all or a portion of a student’s tuition. Research assistantships provide a stipend and require students to work on research projects with faculty in the department for up to 15 hours per week.

Research assistantships are available during the summer. Students in good academic standing who wish to be supported during the summer should notify the DGS early in the spring semester.

Doctoral students in good academic standing can expect five years of funding. Full-time Master’s students in good academic standing can expect two years of funding.

Doctoral students who receive partial or full support from external sources must provide documentation annually affirming their support and detailing its terms. Sources of external support include support from universities, governments, foundations, and corporations.

4.6  Expectations for Supported Students

Students who receive merit-based scholarships, assistantships, or both through the department must maintain high academic standards, maintain satisfactory, sustained progress toward the completion of their degree requirements, and participate in the academic life of the department. To remain eligible for departmental support, Master’s students must maintain a 3.78 GPA and doctoral students must maintain and complete their course work with a grade-point average of 3.5. Criteria the department uses to determine merit include but are not limited to academic standing, enrollment status, attendance of departmental seminars and colloquia, faculty assessment, participation in faculty research, scholarly publication, scholarly presentation, and attendance of scholarly meetings, such as conferences. For students who hold assistantships, if they accept employment outside of the department, they must report this employment to the DGS, and it must be limited to eight hours per week.

The department expects supported Master’s students to complete all degree requirements before the end of their second academic year. The department expects supported doctoral students to complete all degree requirements before the end of their fourth year, but no later than the end of their fifth year. Furthermore, the department expects supported doctoral students to complete their Master’s course work before the end of their second academic year, pass their qualifying exam and defend their dissertation proposal before the end of their third year, and successfully defend their dissertation before the end of their fourth year, but no later than their fifth year. Supported students who obtain advanced standing because of a previous graduate degree are subject these same guidelines, although they will be adjusted based on individual circumstances.

4.7  Failure to Meet Expectations

The department expects all students to maintain high academic standards, make satisfactory, sustained progress toward the completion of their degree requirements, and participate in the academic life of the department. Students who do not meet the expectations and standards set forth in this handbook and in the Graduate Bulletin may be informed that they are not making satisfactory progress toward their degree, may have their financial support reduced or eliminated, or may be dismissed from the program. Students having difficulty for any reason should immediately contact their professor, advisor, mentor, or the DGS, since waiting could eliminate the possibility of mitigative or corrective action.

At the end of each semester, the department reviews the progress and standing of all students in its graduate programs. Students who have concerns about their standing or progress may submit a one-page statement to the DGS before the end of the final-exam period that details their recent activities and progress in the program.

Students who are not meeting expectations, depending on the duration and severity of the situation, may receive an oral or written warning from the department, a written warning from the Graduate School, or a letter of termination from the Graduate School.

Students who earn unacceptable grades or whose GPAs fall below the required minimums for departmental funding or for good academic standing are placed on probation for one semester. The department’s Graduate Committee communicates specific academic goals for the probationary semester. Failure to meet these goals may result in another probationary semester or termination from the program. When it becomes mathematically impossible for students to raise their GPAs above required minimums, they are ineligible for funding or may be dismissed from the program.

For the first unacceptable grade students earn, they receive a written warning from the Graduate Committee; for the second, the Graduate Committee reviews student’s overall standing and progress, and may recommend dismissal from the program. (Master’s students should earn no grade lower than C; doctoral students should earn no grade lower than B-.) Students cannot retake courses (unless they receive a failing or unsatisfactory grade), and they cannot take additional courses beyond those required for the degree in an effort to raise their GPA above a required minimum.

Doctoral students ineligible for department support must obtain support from a member of the faculty, unless they have support from a legitimate external source. Doctoral students without such support will be dismissed from the program. Doctoral students can not “self-pay.”

4.8  Success in Exceeding Expectations

The department expects all students to maintain high academic standards, make satisfactory, sustained progress toward the completion of their degree requirements, and participate in the academic life of the department. Students who exceed the expectations and standards set forth in this handbook may have their financial support increased or may be eligible for an award or other recognition.

At the end of each semester, the department reviews the progress and standing of all students in its graduate programs. Students who have significant accomplishments to report are encouraged to submit a one-page statement to the DGS before the end of the final-exam period. Such accomplishments include but are not limited to departmental recognition, university awards, best paper awards, and external fellowships.

4.9  Departmental Talks

When graduate students publish papers in top journals and conference proceedings in their area, they should consider giving a talk in the department. Such talks could be useful in advance of a presentation at a conference. Students and their advisors should work with the Colloquium Chair to schedule departmental talks.

4.10  Outside Courses

Students who wish to take courses outside of the department and the approved external electives must apply in writing to the DGS and obtain approval before registering for the course. This includes Consortium Courses and courses that may be eligible for transfer, such as non-degree courses taken at another university over the summer. Students should provide course descriptions, syllabi, and any other information about the courses that will help with the review. The DGS reviews petitions for outside courses in the same manner that petitions for transfer credits are reviewed.

4.11  Transfer of Credits

Master’s and Ph.D. students can apply to transfer up to three graduate-level courses (i.e., 25% of the required course work) from another institution toward the MS requirements provided that the courses have not been used for an earned degree. Students must have earned a grade of B or better in the courses.

After the first semester of full-time registration, students may apply for transfer credit through the DGS by completing and submitting the Graduate School’s form Student Petition for Change to Graduate Program. Students should also provide course descriptions, syllabi, unofficial transcripts, and any other information about the courses that will help with the review. Students must also ensure that the Graduate School has official transcripts for all courses.

After receiving a petition, the DGS contacts a faculty member qualified to evaluate the petition and make a recommendation to the DGS. The DGS makes a decision based on the faculty recommendation. The DGS communicates the decision to the student, and if it is positive, then the DGS forwards the decision as a recommendation to the Graduate School. If the Graduate School accepts the recommendation, then it records the grade and credits on the student’s transcript, but the grades do not count toward the student’s grade-point average.

4.12  Advanced Standing

Doctoral students who have earned Master’s degrees can apply for advanced standing. After students have been accepted to the doctoral program and have indicated their intent to matriculate, the DGS contacts eligible students before the start of the fall semester and explains the process for applying for advanced standing.

4.13  The Timing of Reviews and Defenses

Graduate students writing theses or dissertations must understand and take into account the overall time required for the review of the documents required for the degree requirements. Committees have at least three weeks to complete their review, but the review could take up to five or six weeks, especially if members are away from campus or if there are problems with the draft. Modifications to a document after a review could require one or two additional weeks. It is reasonable for students to assume that the members of their committee will be available and responsive during the academic year, but they may be less available and responsive between semesters and during the summer. Indeed, faculty members are under no obligation to review documents or attend presentations during the summer.

Once an advisor informs a student to produce the final draft of the thesis, proposal, or dissertation, the student should contact the members of the committee and determine their availability for the current or forthcoming semester. If a final draft of a thesis is well-written and requires only minor modifications, then students may be able to submit it to their committee before the midpoint of the semester in which they plan to defend. If a final draft of dissertation is well-written and requires only minor modifications, then students may be able to submit it to their committee near the beginning of the semester in which they plan to defend. The extra time required for the defense of the dissertation takes into account the additional complexity and length typical of dissertations and the additional requirement of the pre-defense. It is important to keep in mind that these are best-case time lines. The review of some theses or dissertations could require more time.

The best strategy for increasing the likelihood that the review proceeds in a timely manner is to make sure the final draft of the thesis or dissertation is well-written and ready to defend before submitting it to the advisor and committee. Naturally, the advisor is responsible for helping the student make this determination. Additionally, students should communicate with their committee throughout the research project, but communication with the committee is particularly important during the period leading up to the review and defense.

4.14  Defending during the Summer

Students must petition to defend their thesis or dissertation during the summer, defined as the period between the Graduate School’s last date to defend for the spring semester and the start of classes of the following fall semester. The petition must consist of letters of support from the members of the committee, a time line of remaining milestones, and a current draft of the thesis or dissertation. The Graduate Committee approves the petition with a majority vote. The petition must be submitted to the DGS before April and is in addition to any petition required to extend the time for completion of the program.

4.15  Requests for Extension of Time Limits

Students may request for extensions of time through their advisor, the DGS, and the Graduate School using the Student Petition for Change to Graduate Program. See the Graduate Bulletin for more information about requests for extension of time limits.

4.16  Optional and Curricular Practical Training

International graduate students may participate in off-campus employment through Optional Practical Training (OPT). The department makes no arrangements for MS students to participate in Curricular Practical Training (CPT). Doctoral students with the approval of their dissertation advisor may participate in CPT during summer semesters by registering for Thesis Research after they have completed their requirements for the MS program.

To qualify for post-completion OPT, course-work Master’s students must complete all of their courses. Master’s students writing a thesis must pass the review with only minor revisions and schedule a date for their defense. Doctoral students must pass the pre-defense and schedule a date for their defense.

Graduate students who plan to participate in OPT while employed by the department must first obtain permission from their academic advisor and the DGS. The student, advisor, and DGS must develop a work arrangement for departmental duties given the additional outside employment.

4.17  Outside Employment

Doctoral students and graduate students employed by the department must limit employment outside of the department to no more than eight hours per week. Students who intend to pursue employment outside of the department, such as consulting or teaching, must first obtain permission from their academic advisor and the DGS. (International students should consult the previous section on Optional Practical Training.)

4.18  Application to Graduate

Within the first three weeks of the semester that they intend to complete the requirements for the degree, students should apply to graduate and notify the DGS of their intent. Note that students who plan to attend commencement ceremonies in May must apply no later than February 1 to have their diploma available and their names printed in the program.

4.19  Students in Other Georgetown Graduate Programs

Graduate students in other departments or programs can take graduate courses in computer science or undergraduate courses in computer science for graduate credit in their home department. Graduate students in other departments or programs can take up to two graduate courses in computer science with permission from their advisor or DGS. Taking more than two graduate courses in computer science requires a petition to the DGS of the Department of Computer Science. The petition must state how the classes support the student’s plan of study and must include a letter of support from the student’s advisor or DGS.

Graduate students in other departments or programs can take undergraduate courses in computer science for graduate credit in their home department. The home department must work with the Department of Computer Science to create a cross-listed graduate-level course in the home department for an undergraduate course in computer science. For the cross-listed course, an instructor in the home department must be the primary instructor of record and the instructor of the undergraduate course must be the secondary instructor of record. The Department of Computer Science allocates seats for the cross-listed graduate course, and the home department manages the enrollments in the course. The instructor in the home department is responsible for making the material covered in the undergraduate course relevant to their discipline and for defining and grading the additional assignments required for graduate credit. The students in the home department enroll in the cross-listed course, participate in and complete all of the assignments required for the undergraduate course, and complete all of the additional assignments required for graduate credit. At the end of the semester, the instructor of the undergraduate course provides the student’s grades and any other pertinent information to the instructor in the home department, who assigns the final grade for the graduate course.

4.20  Departmental and Graduate-School Deadlines

For any Graduate-School deadline, unless specified otherwise, the department’s deadline is one week prior.

5 Appendices

A. Changes to Previous Version

• Added:
  • rescheduling qualifying exams
  • postponements of qualifying exams
  • externally supported doctoral students
  • doctoral students and CPT
• Modified:
  • eligibility for departmental support
  • procedures for qualifying exams

B. Previous Versions

• August 27, 2012
• January 20, 2012
• January 9, 2012
• August 29, 2011