EECS 545 Project Guidelines

Every group who has not already done so needs to meet with me on Nov. 2 or 3 to discuss the project topic. One person from each group, send me email stating which 30 minute intervals your group is available, Wed. 1 to 5 pm and Thurs. 2 to 5 pm. By the time of this appointment, you need to have identified a topic and given it some preliminary thought.

Your project should be an open-ended investigation into some aspect of machine learning. Every project should make some original contribution to machine learning methodology, the application of machine learning to real world problems, or (preferably) both. Projects will be graded based on originality, execution, and the quality of the poster and written presentations. Your project does not need to be publishable research, but you cannot simply redo what someone else has already done. There should be some degree of exploration and innovation.

When choosing your topic, remember this: if you aren't interested in the topic, you won't be motivated to work on it. So choose a topic that captures your imagination!

Project ideas:

To get ideas for a project, you might look at recent publications such as Journal of Machine Learning Research, and the conferences ICML and NIPS. You may also wish to look at past projects, available through the course webpage. Email me in advance of your meeting if you would like help finding a topic, and I'll help if I can. Just supply me with as much information as you can about your preferences.

Here are some topics in machine learning that may be of interest, in addition to those covered so far in lecture:

• Active learning
• Semi-supervised learning
• Novelty/anomaly detection
• Missing data
• Predictive state representations
• Nonlinear dimensionality reduction
• Multi-task and transfer learning
• Online learning
• Ranking and recommender systems
• Sparsity and high dimensional inference

Deliverables and grading (click on links for more info):

• Project concept (5%): Due Nov 7 in class. See HW 4.
• Detailed proposal (20%): Due Nov. 16 in class.
• Final report (50%): Due Dec. 16, at noon. Slide a copy under my door.
• Poster presentation (25%): Please reserve Tuesday, Dec. 20, 3-7 PM, for the poster session. Location: Tishman Hall.