Instructor

• Mingyu Ding
• Email: md@cs.unc.edu
• Classroom: SN011
• Time: Mon/Wed, 10:10–11:25 AM

TA & Office Hours

• Yunchao Yao · yunchaoy@cs.unc.edu
• Office Hours: Wed 1:00–2:00 PM at Robotics Lab (By appointment)

Prerequisites

• Fundamental deep learning concepts (COMP 664 or 755)
• PyTorch / Simulators (Isaac/MuJoCo/SAPIEN/etc.)

Course Description

Robot learning lies at the heart of AI & robotics—teaching robots to make complex, sequential decisions in diverse environments. This graduate-level course blends lectures with paper readings covering core ideas from machine learning, deep learning, vision & language, behavior cloning, and decision-making for control. We will explore methods such as behavior cloning, multimodal perception, and policy learning, while examining open challenges in embodied intelligence and connecting insights to research directions.

Grading

• Class Participation: 20%
• Paper Presentation: 30%
• Course Project: 50%

Course Policies

• Late Submissions: All assignments are due at 11:59 PM ET on the specified date. Only on-time submissions are accepted.
• Academic Integrity: External resources are permitted with proper acknowledgment. See the UNC Honor Code for details.