COMP 875 Project or Survey Paper

Proposal document due: Tuesday, September 29th

Submission: Email me a PDF document (around two pages)

Survey Proposal

The survey paper can be, but does not have to be, on the same topic as your presentation. The final survey paper should be about 15 pages in length and should be structured like a formal academic paper (typeset in LaTeX, single-spaced, 10 or 11 point font, one-inch margins). It should be a self-contained and comprehensive introduction to a machine learning topic that would be accessible to anybody in the class. The proposal should include the following components:
  • Topic description: define the scope of your survey paper. Describe the general research area you plan to cover, along with your personal goals, i.e., what you plan to learn, what questions you want to answer. If the topic area is related to your research area, you should make it clear what you will learn that you have not already learned in the course of your research.
  • Outline of proposed paper.
  • Reference list: this should be ten or so papers at this stage.

Implementation Project Proposal

The goal of the project is to implement one or more learning approaches and to explore their behavior on realistic data. Given the time limitations of the course, please try to define your project as specifically as possible! As part of your project, you may use any code or executables that are freely available on the web, as long as your project goes beyond that code and implements something sufficiently significant on top of it or modifies it in some interesting way. The project proposal should contain the following components:
  • Problem definition: Identify the paper(s) you plan to follow and what you propose to implement. If you have any specific hypotheses you intend to test, list them here. If your project is related to your research area, specify how this project goes beyond your regular research work.
  • Outline of proposed work: Indicate the priority order of different parts you propose to implement. List potential improvements/ extensions you may want to do if you complete the core of your project faster than anticipated.
  • Resources: Specify what data you plan to use. Also specify whether you plan to use any outside code and how you plan to build on it.
  • Potential issues: Try to anticipate which part of the implementation or testing may prove the most difficult. Possible stumbling blocks shouldn't necessarily prevent you from attempting a more ambitious project, but you should talk to me early on to make sure that you can still define the project in a satisfactory manner.