Much of this 'syllabus' applies to undergraduates and graduate students. But, there are a few things specific to undergraduate researchers. Below is my attempt to explain them.
Following much other text in this set of documents, I expect in-person attendance from all students I advise. Not getting much "work done" is not an acceptable reason to cancel meetings, in my view. There is always a lot to talk about regarding your work. If there isn't, you don't need me to advise you!
- I do not fund students to work remotely in my group (e.g., over the Summer).
- I do not hold regular scheduled meetings with unfunded remote students.
You can be paid a modest sum as an undergraduate researcher. This comes in the form of funding routed through either
- A federal grant (via Spencer)
- A salary award from Georgia Tech, called PURA Salary.
I generally expect you to apply for PURA Salary before requesting funds from a sponsored grant. Some example proposals for the salary award are located on the Box. Deadlines are often early, about 3.5 months before the semester of the award. For example:
SPRING 2024 APPLICATION DEADLINE
FRIDAY, September 29, 2023
Details on the review rubric for PURA Salary are found here
I do not fund undergraduate researchers during their first semester or year working with the group. This is primarily due to the start-up time required to get familiar with any project and contribute more directly. The same reasoning applies to why I don't accept undergraduate researchers graduating within two years. Conducting scientifically impactful research takes time (see below).
Note: Please read The When and Where of Working
Following the above:
Conducting scientifically impactful research takes time.
I expect undergraduate researchers in the group to take fewer classes than they would otherwise. This amounts to, at most, 4 serious classes you are conducting research per term. Taking fewer classes might mean graduating in 4 years instead of 3 (or something akin to this). You must consider this trade-off on your terms, though I'm glad to discuss it with you. GT CS offers an undergraduate thesis option, which meets several degree requirements but is otherwise a small time commitment (and teaches you useful things!).
Some commitment topics to consider:
- If you hope to go to a top graduate school (or are considering it), at Georgia Tech or otherwise, completing your degree early is unlikely to help with admission, but completing high-quality research definitely will.
- If you don't think you do not want to go to graduate school, research can still expand industry opportunities. The degree to which it expands those opportunities will depend on the job you hope to obtain. Discuss with Spencer for more details.
The goal of performing research, in part following the above, is to produce scientifically impactful work. As such, one should aim to produce work that one "owns" that was not primarily conducted by someone else. One can argue that this alone is an intrinsic good.
Still, that work should be worthy of a top conference or journal venue (the latter is more typical of our work). Such papers are tangible outputs and, in part, serve as a stamp from a research community saying, "You conducted and communicated scientifically meaningful and important research." Without such a "stamp," it is harder to convince others (future employers, for example) that your work is important to the community. Strong letters of recommendation are great. First-author journal papers are often much better (if you are an undergraduate). The above two things are often strongly correlated.
It takes about two years to research and write a journal paper, plus additional months for review and publication.
If you are interested in continuing to a Ph.D. at GT or elsewhere, there are a few extra things worth noting.
- Having a journal or conference paper(s) in top venues will strongly improve your chances of admission.
- Having extremely strong references, even just one(!) (in this case, likely me), can get you into a top Ph.D. program. I will write a letter faithful to your work with me and the group.
Should you get a Ph.D.? This can be a tough question and is quite individualized. You should talk about it with me if you are thinking about it.
If you want to participate in an internship, I will gladly support you in your search. I can guide you more closely if you want your internship to work well with your research. Whether you should or should not take on an internship will depend on several factors, depending on your goals upon graduation and your research area. I recommend you talk about this with me so I can give more personalized guidance.
I cannot guarantee a desk in CODA S1347 (or elsewhere), especially as space seems tight (according to the school). However, I try to keep at least one "Bryngelson Group" desk that floats, and you can use it if no one else is (in CODA S1347). You are always welcome to use the CODA spaces on the 13th floor or otherwise. These areas seem quite pleasant to me, but you may disagree. If you would like somewhere to work but are still looking for a comfortable place, please let me know.
Prioritize being in proximity to other group members. Your colleagues are there to help you. Using them effectively, whether you are a UG or not, and whether they are a UG or not, betters both of you and will undoubtedly accelerate your research. If you are not making consistent contact with your colleagues, I will notice and suggest you do so.
If you have made significant research progress, you can present it at a conference! I have sent several undergraduates to conferences; nearly all UGs that worked with the group for > 1 year. You should apply for PURA Travel.
Applications are to be submitted no earlier than three months before the anticipated conference date and no later than one month before the anticipated conference date.
It is a rather short proposal that is easy to put together. Some example successful proposals are here. Undergraduates do not need to complete a Georgia Tech Spend Authorization (but graduate students do).