Introduction to Machine Learning

Course Project Information

The course project is an opportunity for you to explore an interesting machine learning problem of your choice in the context of real-world data. Projects can be done by you as an individual, or in teams of two students. The instructor will consult with you on your ideas, but of course the final responsibility to define and execute an interesting piece of work is yours. Your project will be worth 20% of your final class grade, and will have 3 deliverables. See the syllabus for dates. Note that all final write-ups must be in the form of an ICML paper (8 pages maximum in ICML format, including references). The page limit is strict! Write-ups over the limit will not be considered. Latex and Word templates for AAAI format can be found here. Project proposals have no formatting requirements.

What Makes a Good Project?

There are typically two types of projects: (1) A student is involved in ongoing research, perhaps for an MS of Ph.D. degree, and identifies some aspect of the problem on which they are working for which ML techniques could be useful. (2) A student has some outside interest, such as playing Texas Hold'em or the stock market, and identifies/obtains data from that domain that can used to test the utility of one or more ML algorithms. For example, one could (in theory) use reinforcement learning to learn to play Texas Hold'em, or Support Vector Regression to learn to predict the closing price of IBM stock. In some cases, students identify a research question that is inherently interesting for ML researchers and work on that question. The last time I offered this course, one of the students did a project of this type that was ultimately published at AAAI.

Project Proposal

You must turn in a brief project proposal (1 page maximum). Proposals should include the following information:

Final Report

Your final report is expected to be a 5-8 page report. It should roughly have the following format: