Instructor: Prof. Aleksander Madry (email@example.com)
Schedule: MW2:30-4, room 37-212
While deep learning techniques have enabled us to make tremendous progress on a number of machine learning and computer vision tasks, a principled understanding of the roots of this success – as well as why and to what extent deep learning works – still eludes us.
This course will begin with background lectures, and then shift into a seminar format in which students will learn and give presentations about fundamental ideas and phenomena that underlie recent developments in deep learning. Each presentation will be followed by a class discussion of the merits and shortcomings of the state of the art. Class projects will be designed and carried out with the goal of addressing some of these shortcomings. The focus will be on building a principled understanding of deep learning via a mixture of empirical evaluation and theoretical modeling.
Enrollment may be limited.