Personalized Machine Learning

Grading

30% presenting & critiquing portions of weekly readings

30% homework (background overview/comparisons/implementation of related Machine Learning techniques)

40% final project

There will be no exams. There will be weekly reading and presentation assignments and a final project. All students are involved in two weeks of presenting work - in one week as the presenter/supporter, and in one week as the critic/attacker. Thus they need to learn to take both sides in understanding the work.

Students will be working throughout the semester on one project in small groups (2-3 students max). This will involve: formulating a research question, data organization, model development, testing, and analysis of findings. In addition, students will be assigned implementation tasks (to compare 2-3 related Machine Learning approaches), and will be assigned readings and other exercises to be done between classes. Classes will also include reading seminars, presentations from invited speakers, and continuous evaluation and discussion of the projects’ progress.

Late Policy

Students who miss more than three classes will fail the process component of the class (30%). Exceptions will only be granted with a doctor or MIT Student Support Services note or for religious holidays (advance notification required). Please note that even if an exception is granted, the grade may be affected. However, the final class project presentation attendance is required, both days - it is a key part of the learning experience. Homework that is handed in late will have 10% deducted from the homework grade for each hour of delay after the first hour.