Description

This course will cover a variety of machine learning approaches that allow robots to learn manipulation tasks from their own actions and experiences, as well as through interaction with humans. Topics will include methods from a) imitation learning, b) learning from demonstration, and c) Reinforcement Learning. We will discuss methods including, but not limited, to data gathering and pre-processing, skill encoding, reproduction, and generalization, skill refinement, obstacle avoidance, symbol grounding, symbolic planning, feature selection and segmentation, and active learning. The course includes student presentations and a final project where students develop an existing approach and extend it further by applying and implementing their own ideas. There are no formal pre-requisites however, this course covers material that utilize a good deal of machine learning and there will be no time to cover all the background material. Therefore, I strongly recommend having a graduate-level machine learning course (COMP.5450), equivalent research experience, or the willingness to do significant studying outside of class. Students are also expected to have fair knowledge of (a) Linear algebra, (b) calculus, and (c) statistics.