This is the project proposal for the final project. Proposals should be for projects that are some combination of the following three forms
The project proposal should be 2-3 pages long, and type-set in LaTeX in an appropriate conference format (e.g. IEEE conference template).
You may choose to work on the project either alone or with a partner. If you work with a partner, both of you are still expected to contribute to all parts of the project, including the paper and presentation.
Note: there should be interesting mathematical content in your project. This means that I expect you to do more than take publicly available data and throw it into an existing machine learning pipeline! Although those kinds of projects are perfectly adequate in some conferences, they are not suitable for this class.
I encourage you to take on an ambitious project; if you propose an ambitious project that doesn't work out, you can either (1) fall back to doing an interesting synthesis/review of your chosen topic, or (2) write it up as a negative result, describing why the approach you tried failed and what you might do differently the next time. I will not penalize you for failing an ambitious project, so long as you put in a good-faith effort, learned something from the process, and can articulate it in paper/presentation form.
You are encouraged to come up with your own idea; this is a graduate topics course, so one hope would be that the material you learn here would be relevant to your own research interests. I am happy to chat about project ideas, and will certainly give feedback on the proposal if I think the project is unsuitable for some reason.
To help you out, I've also included some examples of appropriate projects on Quercus on the PS3 assignment page, though I should warn that some of them tend to be on the ambitious side. For most of those projects, I would try to do something novel, but expect that there is a reasonable likelihood that it'll just turn into a literature review by the end of the semester.