PhD Thesis Proposal
After passing the area qualification process you must form a thesis committee and defend a thesis proposal.
Students must perform research that is a significant contribution to the field during their third year. This can be satisfied by:
- Writing a paper that is accepted in a respectable refereed conference or journal
- Producing a paper of similar quality (quality of paper judged by the dissertation advisory committee)
- Incorporating the contribution in the required thesis proposal
Dissertation Advisor and Preliminary Advisory Committee
Soon after passing the area qualification process, you should concentrate on narrowing down your interests to more specific ideas, such as:
- “Truth Maintenance in Natural Language”
- “Collapsing Complexity Classes via Counting”
- “Parallel Visual Shape Recognition”
- “Latency Tolerance in Distributed Shared Memory Systems”
Part of this process will be exploring ideas with faculty and finding a dissertation advisor and a preliminary advisory committee.
All students must register their dissertation advisor and a preliminary advisory committee with the graduate coordinator no later than December 31 in their third year.
Your advisor will play a major role of guiding you through the process of completing a PhD. Your advisor will:
- Help you in planning your thesis proposal defense
- Point you towards to appropriate literature
- Advise proposal-related (and other) research
- Read drafts of your proposal
- Giving general advice
The advisor also plays a crucial role in the actual exam itself. Choosing an advisor should not be done lightly; changing advisors can significantly delay completion of your studies.
- Your dissertation advisor
- At least three University of Rochester faculty members holding the rank of at least assistant professor
- Three department members*
*This is a department requirement. Exceptions can be granted by the chair.
A faculty member from outside the department can also be included, and must be included when the final dissertation advisory committee is formed in the second term of the third year.
After choosing an advisor and a general category, the next step is to decide what you really want to do. This involves finding, with the help of your advisor, a suitable topic.
After choosing a topics students should search through literature to answer the following questions:
- What (if anything) has been done already?
- What has not been done?
- What are the major gaps in previous work?
- What are recognized “next steps”?
After you have a grasp of the area and the problem, you will need to outline how your research will address the problem. This outline should include ideas on:
- How the research will attack the problem
- What it will not attack
- How it will fit in with previous work
- What the essential contribution of the work will be
You should be actively engaged in research on the topic by the fall of your third year.
Dissertation Advisory Committee
Your preliminary advisory committee members will usually become your dissertation advisory committee. If your preliminary advisory committee had no outside member, you must bring one on board at this time.
The committee members should be Rochester faculty members holding the rank of at least assistant professor, and three should be from the Department of Computer Science. (For exceptions, see the section above on forming a preliminary advisory committee.)
Each member must sign your thesis proposal defense form immediately after the thesis proposal defense. Your advisor should promptly return this form to the graduate program secretary.
Producing a Thesis Proposal
This proposal should explain:
- The context of the problem
- The problem itself
- Previous approaches
- Your proposed research
You should also include a well-researched bibliography. The thesis proposal should be of high quality in style, content, and exposition.
The thesis proposal and all other publications you have written during the year should be distributed to the dissertation advisory committee at least ten days before your thesis proposal defense. Students should ideally distribute materials before even scheduling the defense.
The thesis proposal will usually describe your:
- Third-year research
- The specific research directions you will pursue in the immediate future
- The general research directions you will pursue in the more distant future
- The theme that will unify your research into a coherent PhD dissertation
The thesis proposal should demonstrate that you have acquired the skills needed to perform dissertation-quality research. You are expected to have performed new research of substantial strength and novelty since your area paper. Except in exceptional cases, this new research should be appropriate for inclusion in the dissertation.
The thesis proposal should demonstrate that you have the technical strength needed to do PhD-quality research, and the vision to see the “big picture” into which that research fits.
Furthermore, the thesis proposal should show that you not only know how to solve problems, but also how to frame the issues.
Finally, the thesis proposal should demonstrate that you have developed strong and insightful intuitions as to which research themes are promising. The thesis proposal defense serves to verify these points.
In short, the proposal, talk, and exam should demonstrate to the dissertation advisory committee that an entire dissertation is indeed likely to result within a reasonable time frame.
A successful thesis proposal is not a guaranteed formula for producing a successful dissertation. As the research progresses, the research goals may change dynamically, and some initial goals may be too hard to be solved within the time frame.
We therefore expect that the dissertation project will evolve to meet these contingencies, and that this evolution will be the primary topic of six-month reviews.
Scheduling the Thesis Proposal Defense
Once sufficient feedback on the thesis proposal has been gathered, you can schedule the Thesis Proposal Defense. This is best done early in the spring of the third year, though it can be done earlier, and must be done before the spring PAS.
When you are ready to schedule the thesis proposal defense, see the graduate program secretary to reserve a room and date, and to complete a Thesis Proposal Defense Appointment Form.
The graduate program secretary will not schedule more than two events in the same day—one in the morning and one in the afternoon—to ensure the availability of interested faculty members. Students should try to schedule events well in advance to make sure they meet the spring PAS deadline.
Defending the Thesis Proposal
A public presentation is a required part of the thesis proposal defense. It is a chance for you to publicly present your ideas to the community and for your committee to judge both the ideas and the presentation.
The presentation should take no more than an hour, and should concentrate on the proposed research and the current year’s research progress.
You should provide the department secretary with the date, time, place, and abstract of the talk at least ten days in advance. She will then advertise the talk to the faculty, staff, and students.
The actual exam, which will normally occur immediately following the public presentation, is a meeting of the dissertation advisory committee and the student. Other faculty may attend and freely question and comment.
The purpose of the exam is for the committee—now that it has read the thesis proposal and heard the public talk—to ask you further questions and give you feedback.
Questions may address any aspect of the proposal, including the actual research, the larger problem, your familiarity with previous work, and your expected attack on specific sub-problems. In addition to direct feedback, the committee will also report to the PAS.
You may choose to attempt the third-year process in your second year. You will be expected to do so if you passed the area qualification process during your first year.
There are no delayed requirements in this case; accelerating simply amounts to completing the third-year hurdles one year early.