This offering of individual studies in machine learning is a “wrapper” around a COURSERA course on the same topic, adding readings, discussions, and an end-of-the semester project (with a faculty mentor) to the online course offering.
In particular, there will be several components to this individual studies offering during Fall 2012.
1) Students will enroll in the COURSERA course on Machine Learning to begin August 20 and running for 10 weeks (https://www.coursera.org/course/ml ). In part, I think of the COURSERA course content as the (multimedia) “textbook” for this CS 390, but it is also more than that. To receive credit for CS 390, students must view all lectures, do all programming assignments, and take all quizzes for this online course, and submit a certificate of completion from the course instructor, Professor Andrew Ng (See Notes below). As programming assignments and quizzes are completed, the results will be also given to and evaluated by the CS 390 supervisor (Doug Fisher)
2) Students will meet, as a group, with Doug each Wednesday from 5:00 – 6:30 PM during Fall 2012 to synthesize material from the online course with other materials that are assigned by Doug (see 3). These weekly meetings will not be remedial sessions for the online course – rather, it is expected that students will make full use of the online resources, such as Discussion Boards and wiki, for the online course.
3) One conference or journal paper will be assigned per week, typically material that is not directly covered in the COURSERA course, but which complements it and invites synthesis (e.g., decision and regression tree induction; ensemble learning; relational learning, knowledge-biased learning). Each Wednesday meeting will begin with a short 10 minute quiz on the week’s reading.
4) By October 22, each student will have identified a final project topic, ideally in a domain area in which a Vanderbilt faculty member would have a research interest. By October 29, the online course will have ended, and students will devote the remainder of the semester on a final project, which will consist of applying an off-the-shelf learning system (to include the use of student-coded systems built as part on the online course) to an application domain/task. Weekly group meetings after Oct 29 will be used to discuss the projects that students are working on.
5) On Friday, Dec 7 (Reading Day), each student will present her/his project (task definition and motivation; experimental design; experimental results; related work; future work) in an approximately 20-minute presentation. Presentations will begin about 1:00 PM and continue until all students have presented. Attendance by all CS 390 students is required for the duration of presentations. This session will be open also to other students, faculty and staff of the University.
At about mid-semester, then again at the end of the semester, the Center for Teaching will do an assessment of the course, probably desiring to talk to you as a group and individually.