CSE 254: Seminar on Learning Algorithms

Project Guidelines (2007)

If you are taking 254 for four units of credit, you must do a project in addition to giving a presentation and doing the two assignments.  The project should be at the frontier of current research, but need not necessarily move the frontier forward.  Replicating the results of an innovative recent paper would be a good project, for example.  Projects must be closely inspired by one or two specific recent high quality papers.  

Projects must have an experimental component, but "crass empiricism" is to be avoided.  As much as possible, you should use a high-level programming environment such as Matlab.  If you do not have access to sufficient computing resources for the projects, contact the instructor immediately. 

The schedule for the projects is as follows.  

(1) On Thursday May 10, you should hand in a project proposal.  This should explain explicitly and clearly what you will do.  In particular, the proposal should include:

The proposal should be written in well-organized continuous English, as opposed to just an outline.  Most of its text should be reusable in your final report.  The proposal should be two to four pages long when formatted using the NIPS LaTeX style files.  Since you will have to use LaTeX, you should start immediately.  Using Word is not recommended, because LaTeX is much better for formatting equations and tables, and for controlling the placement of figures.

(2) Start work immediately on your project.  The first phase of the project is especially important.  In this phase, some important tasks include:

These are tasks that can and should be performed mostly in parallel, not sequentially.

(3) On Thursday May 17, you should hand in a revised and (if necessary) extended project proposal.  This must take into account comments received from the instructor and other sources. 

(4) On Thursday May 24, you should hand in a progress report.  This should be two to three pages long in NIPS format.   

(5) On Thursday May 31, you should hand in a draft of your project report.  The report should be polished and should resemble a good submission for a machine learning conference.  You should follow the NIPS instructions, and pay careful attention to these NIPS paper evaluation criteria and these guidelines.  Read, think carefully about, and follow all the principles of good writing in the "Nuts and Bolts" guide to rhetoric by Michael Harvey.

(6) On Thursday June 7, you should hand in the final version of your project report.  This will be graded following a revised version of these grading criteria.  Perfect academic honesty is required.  Note: The NIPS 2007 submission deadline is June 8.  If your report is very good, you may consider submitting it to NIPS.

The final version of your report should be a PDF file in NIPS format that is under one megabyte.  Avoid these common formatting mistakes.  To maximize the readability of your paper, and to minimize the size of the PDF file, make sure that you use only standard fonts.  Also provide an ascii abstract of your report that follows the guidelines here and is 150 to 200 words long.

You are expected to spend about ten hours per week on the 254 project, i.e. about 50 hours in total.  If you are spending much less time, you are not putting in the effort of a full-time student and getting a good grade will be difficult.  If you are spending much more time, you should think more about efficiency and prioritization.While doing the project, remember that winning at research is similar to winning in many other fields of endeavor.
  • Build on an idea that has been successful in previous work.
  • Make the description of your work understandable, attractive, and memorable; pick catchy names.
  • Keep the work simple.  Let the basic ideas shine through.
  • More papers = more ways to have impact and be noticed.
It Takes Soup And a Dream     (Newsweek, March 26, 2001, page 8.)

Susan Runkle isn't just M'm! M'm! Good! She's the M'm! M'm! Best! Her Polynesian pork chops have been named a new Campbell's classic, an honor worth $20,000. Peri asked Runkle to dish on how to win a cooking contest:

  • 1. Polynesian pork doesn't just happen. Work backward by building on a taste you enjoy. (Runkle started with the tangy pork-pineapple combo.)
  • 2. Pick a catchy name. "Use alliteration: red raspberry... something."
  • 3. Keep it simple. Skip recipes "six pages long with chipotle chiles."
  • 4. More recipes = more ways to win. "I sent in a pasta dish, a beef casserole--I call it pepper-pot beef--a Tuscan chicken, a Mexican pizza and a chicken potpie.  Husband's not too keen on fish."