Greetings! I’m Tabitha, or “Tab” for short. I am a roboticist and Ph.D. in Robotics candidate at the Robotics Institute at Carnegie Mellon University. I am a member of the Intelligent Autonomous Manipulation group, headed by Prof. Oliver Kroemer.
My Ph.D. research is in causal robot learning for manipulation. Specifically, I am investigating the interplay between robot perception and control through the lens of causality to learn and leverage the causal structure of manipulation tasks. To this end, my research builds toward a causal robot learning system that empowers lifelong autonomous manipulation in challenging, open-world settings, such as homes, hospitals, and restaurants.
Additionally, I am broadly interested in fundamental robotics problems that have strong real-world impact, such as those I explored through my prior internships with NVIDIA’s Seattle Robotics Lab, Lockheed Martin Space, and Uber ATG.
Prior to starting my Ph.D., I invented, developed, and deployed a vision-based localization system for an underwater robot that is currently used for nuclear reactor inspection. This technology was invented through my M.S. in Robotics research with Prof. Nathan Michael and the Resilient Intelligent Systems Lab. Before CMU, I led the development of multiple software capabilities for safety-critical autonomous systems in the aerospace industry.
I am also a Siebel Scholar in Computer Science (Class of 2017) and a 2022 NCWIT Collegiate Award Honorable Mention recipient.
Beyond my technical interests, I am also interested in working towards greater diversity, equity, and inclusion in both academia and society, including ways in which automation can lead to greater societal advancement.
Ph.D. in Robotics (current)
Carnegie Mellon University
M.S. in Robotics, 2017
Carnegie Mellon University
Graduate Certificate in Artificial Intelligence, 2014
Stanford University
M.S. in Aerospace Engineering, 2010
University of Maryland, College Park
B.S. in Aerospace Engineering, 2007
University of Maryland, College Park
11/27/2023 – Our “Causal-HRI: Causal Learning for Human-Robot Interaction” workshop was accepted to HRI 2024! See you in Boulder, CO and online on March 11, 2024!
10/5/2023 – Thank you to all of our attendees, speakers, and organizers of our IROS 2023 workshop! “Causality for Robotics: Answering the Question of Why” was a great success!
8/30/2023 – Thrilled to share that SCALE, our work for discovering skills using causal learning, has been accepted to CoRL! Huge thanks to my co-authors for making this possible!
5/22/2023 – I have joined Lockheed Martin Space’s Advanced Technology Center for an internship on causal learning and discovery! Looking forward to the summer!
4/28/2023 – Our IROS 2023 workshop proposal was accepted! “Causality for Robotics: Answering the Question of Why” will be held on October 5, 2023 in Detroit, MI!
4/12/2023 – Thank you, UM GENDiR, for a wonderful opportunity to talk about my research and lived experiences in robot and human transfer learning for the GENDiR Seminar Series!
9/30/2022 – Successfully passed my Ph.D. in Robotics thesis proposal! Excited to continue research in causal robot learning for manipulation!
4/13/2022 – Excited to share that I have been selected as a 2022 NCWIT Collegiate Award Honorable Mention recipient for my research in causal robot learning!
4/1/2022 – Our work in creating the 15-996 DEI graduate course has been recognized through the CMU Graduate Student Service Award!
2/17/2022 – Gave a talk to the Intelligent Control Lab at CMU! Thank you ICL for the opportunity to share our work in using CREST for causal feature selection!
2/15/2022 – Our “Learning By Doing” NeurIPS 2021 Competition paper is now live on arXiv and will appear in a forthcoming issue of PMLR!
12/22/2021 – Honored to be selected as a 2022 NCWIT Collegiate Award Finalist for my Ph.D. thesis research!
12/10/2021 – Released the “Learning By Doing” NeurIPS 2021 Competition code to GitHub! Recordings of our NeurIPS session are also available on the LBD website.
10/25/2021 – Thank you so much, Toronto, for the wonderful opportunity to speak about our work in causal structure and transfer learning at the AI in Robotics seminar! [Video]
7/12/2021 – Grateful to be selected as an Inclusion@RSS Fellow for RSS 2021!
7/1/2021 – Congrats to Vicky Zeng! Her recent paper, “Visual Identification of Articulated Object Parts”, was accepted to IROS 2021!
6/1/2021 – Thrilled to help co-organize the “Learning By Doing” NeurIPS 2021 Competition! Excited to bring together researchers and practitioners in control theory, reinforcement learning, and causality!
4/19/2021 – Siddharth Girdhar received a Summer Undergraduate Research Fellowship (SURF) for his proposal on Causal Dynamics Models. Way to go, Sid! Looking forward to collaborating this summer!
4/10/2021 – Vicky Zeng’s IROS 2021 submission, “Visual Identification of Articulated Object Parts”, is now available on arXiv. Best of luck for acceptance!
2/28/2021 – My paper submission to ICRA 2021, “Causal Reasoning in Simulation for Structure and Transfer Learning of Robot Manipulation Policies”, has been accepted! Thanks so much to my collaborators for your hard work – this one feels great!
2/12/2021 – Excited to kick off the pilot course of 15-996: “Diversity, Equity, and Inclusion in Computer Science and Society” as a discussion moderator!
6/29/2020 – The source code for DREAM (my NVIDIA internship work) is officially released on GitHub!
5/4/2020 – Vicky Zeng received a Summer Undergraduate Research Fellowship (SURF) for her proposal, “Visual Identification of Articulated Objects through a Large-Scale Dataset”. Excellent work! Looking forward to working with you this summer with Jacky Liang!
2/14/2020 – Gave a talk at the University of Maryland Robotics Center about my M.S. work in ROV state estimation. Thank you for the invitation to speak at my alma mater!
1/21/2020 – My NVIDIA summer internship work (“Camera-to-Robot Pose Estimation from a Single Image”) has been accepted to ICRA 2020!
11/22/2019 – My NVIDIA summer internship work (“Camera-to-Robot Pose Estimation from a Single Image”) is now publically released on arXiv. It is currently under review for ICRA 2020. Thanks to my collaborators at NVIDIA’s Seattle Robotics Lab, especially my mentors Jonathan Tremblay and Stan Birchfield, for the wonderful opportunity!
12/10/2018 – I have accepted a summer research internship with NVIDIA at the Seattle Robotics Lab, headed by Prof. Dieter Fox.
10/18/2018 – I have joined Prof. Oliver Kroemer and the IAM Lab for my Ph.D. research.
My publications span the fields of robotics science and unmanned systems. Below are my most recent publications in robotics and artificial intelligence. More information about my publication history is available in my CV, including 1 master’s thesis, 2 journal publications, and 3 conference proceedings on aerodynamics of unmanned aerial vehicles (my former area of research).
My publications before 2021 are under my previous name. You are welcome to cite my work from before 2021 using the existing citation (my previous name) or my initials, “T. E. Lee”.
Publications:
T. E. Lee*, S. Vats*, S. Girdhar, O. Kroemer. SCALE: Causal Learning and Discovery of Robot Manipulation Skills using Simulation. CoRL, 2023. Also appeared at the LEAP CoRL Workshop. *Equal contribution.
A. Dahmani*, E. Yiu*, T. E. Lee, N. R. Ke, O. Kroemer, A. Gopnik. From Child’s Play to AI: Insights into Automated Causal Curriculum Learning. IMOL@NeurIPS, 2023. *Equal contribution.
A. Dahmani, E. Yiu, N. R. Ke, T. E. Lee, O. Kroemer, A. Gopnik. Toward Understanding Automated Causal Curriculum Learning in Humans and Reinforcement Learning Agents. IMOL and ICLC, 2023.
S. Weichwald, S. W. Mogensen, T. E. Lee, D. Baumann, O. Kroemer, I. Guyon, S. Trimpe, J. Peters, N. Pfister. Learning by Doing: Controlling a Dynamical System using Causality, Control, and Reinforcement Learning. PMLR (NeurIPS 2021 Competition), 2022.
V. Zeng, T. E. Lee*, J. Liang*, O. Kroemer. Visual Identification of Articulated Object Parts. IROS, 2021. *Equal contribution.
T. E. Lee, J. Zhao, A. S. Sawhney, S. Girdhar, O. Kroemer. Causal Reasoning in Simulation for Structure and Transfer Learning of Robot Manipulation Policies. ICRA, 2021.
The SCALE algorithm discovers compact, interpretable skills using causal learning in simulation. (CoRL 2023)
The Learning By Doing competition explored the intersection of causality, control theory, and reinforcement learning for controlling a dynamical system. (NeurIPS 2021: Competition Track)
FormNet: a deep neural network that identifies real-world articulation mechanisms (e.g., cabinets, drawers, etc) from a single RGB-D frame. (IROS 2021)
Causal feature selection for robot manipulation policies using structural sim-to-real transfer. (ICRA 2021)
Developed an approach for camera calibration from a single RGB image of a robot manipulator. This was my internship project at NVIDIA during Summer 2019.
Developed localization algorithms and capabilities for Uber ATG’s self-driving cars as an intern during Summer 2018.
Using a pan-tilt-zoom camera, a remotely operated vehicle (ROV) was accurately localized within a nuclear reactor, providing greater utility and assurance for reactor inspection operations. This project was my M.S. in Robotics thesis research.
An automated, integrated health monitoring system was developed to improve the efficiency of flight test operations. I led this project while working for Boeing.
Carnegie Mellon University
University of Maryland
(last updated: December 23, 2023)