Joining the Lab
Areas of Interest:
The lab focuses on areas of machine learning for perception and robotics, including (this is a non-exhaustive list) learning with limited labeling (few-shot learning, semi-supervised learning, un/self-supervised learning, etc.), multi-domain/task learning, incorporating structured information for such tasks, network calibration/uncertainty prediction/out of distribution detection, multi-modal fusion, goal-driven perception, and applications to robotics. Our interests evolve though, so please check out our latest publications or just ask!
M.S. and undergraduate students:
Please fill out this form
and we will get back to you. Do this relatively early before the subsequent semester.
I tend to accept a few Ph.D. students every year in machine learning for perception and robotics, focusing on the areas above. See below for how to approach the lab depending on your current state:
If you are currently a student at Georgia Tech, please first look over our research and if interested email me. Include your CV and a succinct statement of your past experience and current interests. I highly recommend contacting me many weeks before the semester starts as opportunities tend to fill up quickly.
If you are applying to Georgia Tech, please follow the application process of the college/school you are applying to. Note that for Ph.D. students there are several choices to select from, including a machine learning (ML) and robotics Ph.D. I hire from several of these. Please make sure to mention my name in your application if you are interested in the lab.
I am not currently looking for external visiting researchers/interns. If you are not currently at Georgia Tech and are not applying for a post-doc, the best way to work with me is by applying to the school (see above).
Feel free to contact me about post-doc positions in the areas of machine learning for perception, focusing on areas listed above. Interest in the intersection of robotics and machine learning is welcome as well. Strong research experience and publication records are expected for top venues in these fields (CVPR/ICCV/ECCV, ICLR/ICML/NeurIPS, ICRA/IROS/RSS, etc.)