CAP5610: Machine Learning Spring 2015

Final Project

Assigned: Thursday, February 19
Proposal Due: Tuesday, March 3 by 11:59pm (via Moodle)
Final Presentation Slides Due: Tuesday, April 21, 10am (via Moodle)
Final Presentation: Tuesday April 21 and Thursday 23, 11am-12:15pm, ECS145
Writeup/Code/Slides Due: Tuesday, April 28, 11:59pm (via Moodle)
Peer Review Due: Friday, May 1, 11:59pm (via Moodle)
Author FeedbackDue: Saturday, May 2, 11:59pm (via Moodle)

Synopsis

For this project, which will take roughly four weeks, you are expected to work in a group of 2-4 people, complete a short research project. You can devise your own project from scratch, or use some project from your lab.

Guidelines [PDF]

You need to undertake a research project with some novelty. What constitutes a "research project?" Well, it must be new, something that no one has published before. Naturally we're not expecting PhD-level research in this amount of time. Following, are some examples of what we have in mind. These are listed roughly in order of least to most ambitious:

An experimental evaluation. Implement one or more existing algorithms and design an in-depth experimental evaluation and comparison that goes beyond what was described in the paper(s). Identify the relative strengths and weaknesses and include this in your report.
An interesting extension of prior work. In most cases, we'd recommend implementing the prior method yourself, rather than downloading implementations available online, as this gives you a better understanding of how the method works (and you can avoid mucking around with some one else's code). But this is not a hard and fast rule--if the extension is very significant you may use available code.
A new application of prior work. Apply a known technique to a new application domain, and evaluate its performance.
Develop a new solution (hopefully better!) to an existing problem.
Pose a new technical problem and solve it. Identify a new problem for which no known solution exists, devise a solution, and implement/test it.

Requirements

Proposal

Each team will turn in a one-page proposal describing their project. It should specify:

Your team members
Project goals. Be specific. Describe the input and output.
Brief description of your approach. If you are implementing or extending a previous method, give the reference and web link to the paper.
Will you be using helper code (e.g., available online) or will you implement it all yourself?
Evaluation method. How will you test it? Which test cases will you use?
Breakdown--what will each team-member do? Ideally, everyone should do something imaging/vision related (it's not good for one team member to focus purely on user-interface, for instance).
Special equipment that will be needed. We may be able to help with cameras, tripods, etc.

Each team must submit a proposal, even if you choose one of the research ideas described below.

Turn in the proposal via Moodle by Tuesday, March 3 (by 11:59pm).

Final Presentation

Each group will give a short (10-12 minute) PowerPoint presentation on their project to the class.

The project presentation will be during the last week of class. We will assign a time slot to each group. In the presentation, you should

provide motivation for your project, explaining why it is important and interesting,
explain your research questions,
provide (preliminary) evidence,
draw (preliminary) conclusion.

You can use the computer projector (e.g. via powerpoint) for your talk. Try to make the presentation interesting (e.g. by including a demo). More details on the logistics follow later.

Your final presentation should be uploaded to Moodle.

Final Writeup

Turn in a web page and a paper-style writeup describing your problem and approach. This .zip file should include at least your writeup (as either a PS or PDF file), the webpage (as HTML or PHP) and the source code of any programs you wrote for your project. Include other files if you feel they are appropriate, but obviously explain their relevance in a readme. You may submit a hard copy of these materials in person if you prefer, but we prefer online submission. Naturally, we will not be around the department at 11:59 PM Wednesday, so if you must turn in a hard copy, make sure you do so early in the day.

Do not be late with your submissions. This is not a homework you can turn in late at the cost of 5 points per day; we are getting together and grading these pretty much immediately, since the final grades are due soon after the deadline. For additional guidance in structuring the report, look at the template structure at template structure Not every project fits into this structure, and you might choose a different structure instead. It should roughly follow the format of a CVPR conference paper, including the following:

title, team members
short intro
related work, with references to papers, web pages
technical description including algorithm
experimental results
discussion of results, strengths/weaknesses, what worked, what didn't
future work and what you would do if you had more time

Code

Turn in your code.

Grading

The projects will be graded in the same spirit as research papers are assessed (though we dont expect you to do original work at the same level). Here is a list of things that we will be looking for:

Originality
Relevance to course
Quality of arguments (are claims supported, how convincing are the arguments you bring forward)
Connection to earlier work
Clarity (how clearly are goals and achievements presented)
Scope/Size (in proportion to size of group)
Significance (are the questions you are asking interesting)

Relative to each other, the proposal will account for 15% of the grade, the presentation for 25%, and the report for 60%.

Feel free to come and talk to us about the various aspects of your project (in fact we strongly encourage you to) so that we can make sure that you are on the right track. Dont forget to have fun while doing it; its meant to be something that you are interested in doing!