Final Project

SCOPE:

The goal of the final project is for you to explore and push the boundaries in unsupervised learning.  This could be in a variety of ways: e.g.  proposal+evaluation of new algorithms / architectures,  investigation of an application of unsupervised learning, benchmarking unsupervised learning, compression, studying synergies between unsupervised learning and other types of learning, etc.

Ideally, the project covers interesting new ground and might be the foundation for a future conference paper submission or product. 

PROJECT TOPICS / STAFF INPUT:

We encourage trying to come up with your own project idea.  We are also happy to make suggestions and/or brainstorm ideas together.

One of the main reasons we are so excited to teach this class is to see more Deep Unsupervised Learning projects happen.  We are very excited to advise on your projects, please don't hesitate to come to office hours to discuss project ideas, project progress, ideas for next steps, etc.

PROJECT TEAMS:

Final projects can be done alone, in a group of 2, or in a group of 3.  Our expectations will scale linearly with the number of people in the group.  We recommend groups of 2 or 3.

PROJECT TIMELINE:

- February 28: Project Proposals Due: 1 page description of project + goals for milestone.  -- Submission through google doc shared with instructors, so we can give feedback/suggestions most easily.

- March 8: Approved Project Proposals Due: by this time your proposals should have incorporated instructor feedback, at this stage it should be assured that your proposal is of right fit and scope

- April 5: 3-Page Milestone Due:  This is to make sure you are indeed making progress on the project and an opportunity to get feedback on your progress thus far, as well as on any revisions you might want to propose to your project goals.  Expectation is that you report on some initial experimental findings (or if you are doing something purely theoretical, some initial progress on that front).  -- Submission through google doc shared with instructors, so we can give feedback/suggestions most easily.

- May 10: 6-Page Final Project Report AND 5-Minute Video Presentation Due:
The Final Project Report should be structured like a conference paper. I.e., focus on the problem setting, why it matters and what's interesting/novel about it, your approach, your results, analysis of results, limitations, future directions. Cite and briefly survey prior work as appropriate but don't re-write prior work when not directly relevant to understand your approach.  References are not counted against the 6 pages.
The Final Project Presentation Video should be structured like a (short) conference presentation.  Start with big picture motivation as well as motivation for the specific topic investigated / method proposed.  Clearly explain the method.  Explain the 1-2 most interesting experiments.