Course Description 

This course will explore various topics in Embodied Artificial Intelligence (AI), which is concerned with the perception and representation of the physical world by autonomous agents and their consequent physical interactions. Specifically, the course will cover how an agent can infer the physical and semantic states of the scene, e.g., via 3D reconstruction and semantic parsing; how these perceptions can be represented in a way that facilitates reasoning with both efficiency and explainability; and how actions can be made in order to achieve downstream interaction tasks. By the end of this course, students will have gained an understanding of the fundamentals of perception, representation, and control in AI systems, as well as learned how embodied agents can be used to interact with and manipulate the physical world.  

This course is designed for students with electrical engineering or computer science backgrounds who aim to perform research in the intersection of computer vision, machine learning, and robotics.   

Prerequisites 

Basic knowledge of linear algebra, probability and random processes, optimization, signal processing, and machine learning will ease the digestion of the course materials.