Course Logistics
Course Description#
This course will cover a wide variety of topics in embodied intelligence, including but not limited to control, reinforcement learning, self-supervised and unsupervised action, visual learning for robotics, generalization and meta-learning. The course will provide an overview of relevant topics and open questions in the area. There will be a strong emphasis on bridging the gap between many different fields of AI. The goal is for students to get both the high-level understanding of important problems and possible solutions, as well as low level understanding of technical solutions. We hope that this course will inspire you to approach problems in embodied learning from different perspectives in your research.
Course Format#
The course will be a mix of lectures and paper readings. Students will be given a large list of papers (both new and old) related to the weekly topics as readings, and will have to sign up to present a paper once in the semester. Weekly lectures will be broken down in the following way: Mondays will be dedicated for the instructor presenting a high-level area in embodied learning, Wednesday will be two students presenting chosen papers on the weekly topic, as well as a group discussion. Students are heavily encouraged to read all the papers, and class participation grades will be based on the involvment in the group discussion. In addition to the presentation, there will be two mandatory papers for students to read for every lecture. Once a week, you need to write 1-2 paragraphs quickly summarizing and discussing the two papers, posting it as a private note to the instructors on Piazza.
In order for students to explore a hands-on area, there will be a class project as well. There will be midterm project updates due as well as final presentations.
Prerequisites#
- Working knowledge of deep learning, reinforcement learning, computer vision and robotics is assumed.
- Experience with reading and understanding academic papers.
- This is NOT a basic course in reinforcement learning, computer vision, robotics, deep learning or AI. This course is meant for you to engage in an interdisciplinary discussion and think about embodied intelligence from different perspectives.
Course Information#
Class Time and Location#
Time: Monday/Wednesday 11:50am-1:10pm
Location: NSH 3002
Office Hours#
| Office Hours | |
|---|---|
| Deepak Pathak | Wednesday 1:10-2:00 PM, NSH 4228B |
| Shikhar Bahl | Monday 1:10-2:00 PM, EDSH 224 |
Class Forum#
Forums are on Piazza (sign up with your andrew email address). Please check out the Piazza regularly, as we will make new annoncements on Piazza. We encourage you to engage in discussions on Piazza as well, which will count towards class participation credit.
Grading Policy#
| Percentage | |
|---|---|
| Paper Presentation | 20% |
| Paper Reviews | 20% |
| Class Participation | 10% |
| Final Project | 50% |
Paper reading summaries count as part of the Class Participation grade and are due Tuesdays AOE before the start of lecture. Late submissions will be penalized.