COMS 6998 Topics in Robot Learning

Overview:

This is an advanced seminar course that will focus on the latest research in machine learning for robotics. More specifically, we study how machine learning and data-driven method can influence the robot’s perception, planning, and control. For example, we will explore the problem of how a robot can learn to perceive and understand its 3D environment, how they can learn from experience to make reasonable plans, and how they can reliably act upon with the complex environment base on their understanding of the world. Students will read, present, and discuss the latest research papers on robot learning as well as obtain experience in developing a learning-based robotic system in the course projects.

  • Instructor: Shuran Song
  • Time: 4:10 - 5:25 MW
  • Location: 327 Mudd Building
  • TA: Zhenjia Xu (zx2282@ office hour 4:00-5:00 Tue/Thur CEPSR 6LW1)

Announcement:

  • Piazza: link
  • Please email your topic choice to the instructor and TA before Friday.
  • Add your project proposal presentation here.
  • [new] Final Project Report Due Dec 16.
      • Instructions on the final project presentation and report here.
      • Presentation link

Topics include:

  • Basic knowledge for machine learning in robotics systems
  • Using simulation environments for robot learning
  • Learning for robot perception (with a focus on 3D perception e.g. 3D object detection, 6D pose estimation, and SLAM)
  • Learning for planning and control (supervised learning, self-supervised learning, reinforcement learning, imitation learning)

Pre-requisites:

  • Familiarity with the basic linear algebra.
  • Knowledge of basic machine learning, computer vision and robotics. Taken any of the following course or equivalent: COMS 4733 or COMS 4731.
  • You need to obtain instructor approval to take the course.

Grading:

  • Topic presentation (40%):
    • The presentation should include an overview of the sub-topic and presentation of related research papers (2-3) on each sub-topic.
    • Prepare slides and drop by my office one week before the presentation to discuss about the slides.
    • For easy sharing and collabrate, please use google slides to make your presentation.
  • Project (group of 1-3) (40%): proposal presentation, final presentation, code, writeup, optional video.
  • Class participation (20%): ask questions and participate in discussions

Policies and Procedures Regarding Academic Honesty

Please read: https://www.cs.columbia.edu/education/honesty/

Schedule:

Click [Paper list] tap below to show the list of papers, and [Schedule] tap to show course schedule.

This schedule is preliminary and subject to change as the term evolves :

Not seeing anything above? Reauthenticate