Graduate Program Requirements
Research Paper Review Ph.D. Exam(registration form)
1. Examination Objectives
The objective of the exam is to assess the student s potential to begin doctoral-level research. The latter requires the student to demonstrate abilities to:
- Read and understand research papers in their field;
- State a problem clearly, provide the motivation and the requirements for a solution;
- Determine if a solution is correct;
- Assess to what extent a presumably correct solution solves the problem;
- Identify potential next research steps;
- Communicate effectively, both in writing and orally; and
- Answer questions relating to the problem and its solution.
The examination is based on two advanced research papers in an area related, but not necessarily identical to the students desired research area. The examination involves a written report, an oral presentation, and general questioning by the graduate faculty.
A student taking the exam will prepare a written report investigating the two assigned papers. The report must give a detailed overview of the papers, present the key ideas, and describe their significance to the research area. The report should not be a mere restatement of the material in the papers. The student is encouraged to criticize, make improvements, suggest alternatives to the approaches described in the papers, and discuss connections between the ideas presented in the two papers. In general, the report should demonstrate the abilities listed in Section 1.
The student will give a 30-40 minute oral presentation of the two papers to the graduate faculty. Following the presentation, the graduate faculty will question the student regarding topics directly related to the two assigned papers as well as general knowledge in this area. After the questioning is completed, participating graduate faculty votes on the outcome of the examination, based on the combined written submission and oral examination. The student is notified of the outcome in a letter sent by the CS&E Graduate Program Committee.
3. Administration of the Exam
Students must submit the completed form by October 1 for January examination and by February 15 for the May examination.
The student, in consultation with the major research advisor, will select the two examination papers. At least one paper must be published in ACM, IEEE, SIAM, Springer, or Elsevier journals; the second paper may be a full-length conference paper or a paper in a journal published by other publishers. The student must submit to the CS&E Graduate Program Committee a written report based on the two papers at least two weeks before the oral examination date.
All members of the graduate faculty may participate to the oral examination and subsequent vote. A minimum of four graduate faculty members must participate in the examination, of which at least one must be a member of the CS&E Graduate Program Committee. Each participating graduate faculty member must vote to either pass or fail the student, abstaining from the vote is not permitted. A strict majority of the votes of the participating graduate faculty is required to pass the exam.
Students entering the Ph.D. program with an M.S. degree in Computer Science, Computer Engineering, or Computer Science and Engineering must take the exam at the end of their 1st semester; all other students must take the exam at or before the end of their 3rd semester. A student who fails on the first attempt may retake the exam once, possibly in a different area and under the supervision of a different advisor. The student must retake the exam at the end of the semester following the first attempt. The papers used when re-taking the exam must be different from the papers used in the first attempt.
Upon failure to pass the second attempt, students will be allowed to continue their studies towards a terminal M.S. degree provided that they are in good academic standing and do not already have an M.S. degree in Computer Science and Engineering from the University of Connecticut. Extensions to the above examination deadlines will be permitted only in exceptional cases, and will require approval of at least half of the graduate faculty.
All students entering the Ph.D. program in the Fall 2004 semester or later will be required to take the exam. For the 2004-2005 academic year, there will be three scheduled examination dates, in January, May, and August 2005. For the following years, oral examinations will be scheduled at least twice a year, normally in January and May.
4. Material from past exams.
- Seth Freeman Presentation Report
- Bogdan Pasaniuc Presentation Report
- Hai Yan Presentation Report
- Haimin Zhang Presentation Report
Related Area of Study
If a related or supporting area is chosen, the courses selected must
comprise a coherent unit of advanced (i.e., 200's level not open to
sophomores or above) work outside the major field of study and
ordinarily outside the department in which the work of the major field
is offered. The courses must be approved by the advisory committee as
a part of the Plan of Study. Ordinarily, they must be taken at the
University of Connecticut. No course credits will be accepted in
transfer toward the related or supporting area unless approved by the
Executive Committee before the courses are taken. With the approval of
the advisory committee, however, the passing of an examination may be
substituted for the course work.
With the consent of the advisory committee, a three-credit advanced
course in mathematics or statistics passed satisfactorily at the
University of Connecticut may fulfill the otherwise six-credit-minimum
requirement if the student's preparation contains a suitably advanced
prerequisite course (i.e., equivalent to a 200's level University of
Connecticut course not open to sophomores) passed satisfactorily at
this or another institution (although no course credits will be
accepted in transfer).
[Back]
Residence Requirement
A graduate student can fulfill the special demands of a doctoral
program only by devoting a continuous period of time to concentrated
study and patient research with a minimum of outside distraction or
employment. During the second or subsequent years of graduate work in
the field, at least two consecutive semesters or, with the consent of
the advisory committee and the student, one semester together with a
contiguous summer period consisting of Summer Session I and Summer
Session 11 of full time study (six credits or the equivalent in each
session) must be completed in residence. This residence period must be
completed at the Storrs campus.
The essential criterion for full-time study as required for
fulfillment of the doctoral residence requirement is whether the
student is in fact devoting essentially full-time effort to studies,
without undue distraction caused by outside employment. It is left to
the advisory committee to determine whether a student's outside
employment is a distraction that prevents the student from devoting
essentially full-time effort to the planned program. The advisory
committee shall record this determination on the Plan of Study
together with a description of the nature, extent, and period(s) of
employment in all cases of approved outside employment during the
residence period.
[Back]
Plan of Study Submission
The Plan of Study must be prepared in triplicate, signed by the
student and the members of the advisory committee, and submitted to
the Graduate Records Office for approval by the Executive Committee of
the Graduate Faculty Council when the student has completed not more
than 12 credits of course work to be applied to the degree. The
student may not take the Ph.D. General Examination before the plan of
study has been fully approved. Failure to present the plan on time may
prolong the period of study for the degree. Before formulating and
signing the plan, the major advisor should have on file and should
consult for guidance a set of transcripts of all undergraduate and
graduate work the student has taken.
Courses elected shall be consistent with the student's objectives and
related to the field in which the degree will be taken. Plans of study
shall consist largely of courses at the 300's level or above. A
limited number of credits at the 200's level (ordinarily not more than
six), if not open to sophomores, may be accepted.
The plan shall designate any foreign language requirement in which the
student is to be tested or any courses comprising a related area of
study. Course credit by examination is not allowed as a means of
accumulating credits to meet the requirements for advanced degrees at
the University of Connecticut. If an examination is permitted to be
used to fulfill a related- or supporting-area requirement for the
Ph.D. degree, course credit is not given. No course credit is given
for the dissertation or for research toward it. Although no credits
actually are assigned, the dissertation is regarded as an important
part of the student's program and is considered to represent at least
one year of full-time graduate study.
Advanced course work taken on a non-degree basis at the University of
Connecticut may be included on a Ph.D. Plan of Study provided the
following conditions are met:
- The grades earned in such course work are B or higher
- Such course work is within the seven or eight year limit (whichever applies) for completion of Ph.D. degree requirements,
- Such credits have not been applied toward any other degree here or elsewhere, awarded or to be awarded.
After approval of the plan by the Executive Committee, any request for change must be submitted to the Graduate Records Office on an official form bearing the signatures of the members of the advisory committee and the student, for approval by the Executive Committee. The successful completion of all work indicated on the approved plan of study is a fundamental prerequisite to the conferring of the degree.
Once approved, the student and the advisory committee should reevaluate the Plan of Study regularly and modify it, following the established procedure, should that be necessary.
[Back]
Ph.D. Publication Requirement
All CSE Ph.D. students are required to publish (or have accepted for
publication), prior to their dissertation defense, a minimum of 3
conference level papers, where each paper is classified as a
peer reviewed full conference article, i.e., submitted and reviewed as
a full paper and not as an abstract. Further, a major advisor has the
authority to establish a higher than 3 minimum number of publications
for his/her students. Ph.D. candidates must make all publications
available (with full citations) in electronic form on the web and
(prior to the defense) and hard-copy form (at the defense).
Ph.D. students admitted in fall 2002 or later must satisfy
this publication requirement.
[Back]
Ph.D. Dissertation Proposal
The Ph.D. dissertation must make a significant contribution to the
computer science discipline. The selection of a dissertation topic and
an advisor to supervise the research effort is the most critical part
of the Ph.D. program. A general area of research is usually selected
during the first year of Ph.D. study and an agreement is obtained from
a faculty member to serve as the major advisor.
After selection of a research area has been approved, an initial
investigation of the relevant literature in the area is undertaken to
establish necessary background information and to define the exact
problem to be studied. This initial investigation may also include a
preliminary experiment to prove the feasibility of any experimental
program that will be included in the research effort.
The initial investigation of the relevant literature in the area will
culminate in the preparation of a dissertation proposal. It is
expected that the student will work closely with his or her major
advisor while preparing the proposal. Acceptance of this proposal by
the student's advisory committee must be obtained before the student
begins the proposed research effort.
A dissertation prospectus of the proposed research, using a special
form obtained at the Graduate Records Office, must be submitted to the
Chairman of the Area Review Committee for Engineering.
The dissertation research usually involves a substantial theoretical
contribution verified by an experimental test of the validity and
applicability of the theoretical results. Purely theoretical or mostly
experimental research efforts are also acceptable provided they make a
significant contribution to the understanding of a given area of
computer science research.
It is very important that the student keep all members of the advisory
committee informed of the progress of the research as it is
performed. This may be done by informal meetings, the presentation of
seminar talks, or the preparation of intermediate summary reports.
The final draft of the dissertation must be presented to the advisory
committee at least one month before the final copy is due in the
graduate school. In writing the dissertation, it is imperative that
the student work closely with his or her major advisor. Experience has
shown that many preliminary copies of the dissertation are needed
before the final draft copy is ready for presentation to the advisory
committee.
[Back]
Ph.D. Dissertation Prospectus
Before preparation of the dissertation is well underway, the student
must file a prospectus of the proposed research, using the special
form obtainable at the Graduate Records Office and following
guidelines included thereon. Failure to file the prospectus early may
result in wasted effort on a dissertation if changes are required in
the project. If human or animal subjects are involved in the proposed
research, the major advisor certifies by signing the above-mentioned
special form that all required institutional and external approvals
have already been obtained and that documentary evidence of these
approvals can be produced by the major advisor upon request.
Once approved by the student's advisory committee, the dissertation
prospectus is submitted to the Area Review Committee for Engineering
which is a subcommittee of the Executive Committee of the Graduate
Faculty Council.
The Area Review Committee for Engineering examines the prospectus on
the following bases:
- Is the prospectus well written, well organized, and well argued?
- Does the prospectus describe a project of appropriate scope?
- Does the student demonstrate a knowledge of the subject and an
understanding of the proposed method of investigation?
- Does the student show awareness of the relevant research by others?
- Does the student consider how the proposed investigation, if successful, will contribute to knowledge?
The prospectus must be submitted to the chair of the Area Review Committee for Engineering at least six months prior to the filing of the dissertation at the Graduate Records Office and must have been approved by the Executive Committee of the Graduate Faculty Council at least three months prior to the filing of the dissertation.
[Back]
Ph.D. Dissertation
The Ph.D. dissertation must make a significant contribution to the
computer science discipline. The selection of a dissertation topic and
an advisor to supervise the research effort is the most critical part
of the Ph.D. program. A general area of research is usually selected
during the first year of Ph.D. study and an agreement is obtained from
a faculty member to serve as the major advisor.
After selection of a research area has been approved, an initial
investigation of the relevant literature in the area is undertaken to
establish necessary background information and to define the exact
problem to be studied. This initial investigation may also include a
preliminary experiment to prove the feasibility of any experimental
program that will be included in the research effort.
The initial investigation of the relevant literature in the area will
culminate in the preparation of a dissertation proposal. It is
expected that the student will work closely with his or her major
advisor while preparing the proposal. Acceptance of this proposal by
the student's advisory committee must be obtained before the student
begins the proposed research effort.
A dissertation prospectus of the proposed research, using a special
form obtained at the Graduate Records Office, must be submitted to the
Chairman of the Area Review Committee for Engineering.
The dissertation research usually involves a substantial theoretical
contribution verified by an experimental test of the validity and
applicability of the theoretical results. Purely theoretical or mostly
experimental research efforts are also acceptable provided they make a
significant contribution to the understanding of a given area of
computer science research.
It is very important that the student keep all members of the advisory
committee informed of the progress of the research as it is
performed. This may be done by informal meetings, the presentation of
seminar talks, or the preparation of intermediate summary reports.
The final draft of the dissertation must be presented to the advisory
committee at least one month before the final copy is due in the
graduate school. In writing the dissertation, it is imperative that
the student work closely with his or her major advisor. Experience has
shown that many preliminary copies of the dissertation are needed
before the final draft copy is ready for presentation to the advisory
committee.
[Back]
