Ran (Thomas) Tian - 田然
rantian [at] berkeley [dot] edu
I am a PhD student at UC Berkeley
advised by Prof. Masayoshi Tomizuka
and Prof. Andrea Bajcsy at Carnegie Mellon
University.
I am also a Qualcomm
Innovation Fellow ,
a World Artificial Intelligence Conference Rising Star ,
and a Robotics:
Science and Systems Pioneer.
My research lies in the intersection of robotics and AI with a focus on
safe alignment between embodied agents and humans.
I tackle the alignment and safety problems that emerge throughout the
life-cycle of foundation models in robotics, ranging from: training (wherein we need to collect
and quantify what kinds of embodied data will enable the desired robotics capabilities), to
fine-tuning (wherein we must align these models with humans), to deployment (where these models
must run in real-time, reliably detect out-of-distribution scenarios, and confidently handover
control to fallback-strategies).
I ground my work through a variety of applications, from autonomous cars, to personalized robots, to
generative AI and
in experiments with real human participants.
During my PhD study, I am a long-term intern at Waymo (2022-Now), working on foundational motion
model for
autonomous driving research (including pre-training, post-training preference alignment, and
distillation for onboard
deployment).
I am fortunate to have the opportunity to intern at the Autonomous Vehicle Research
Group at NVIDIA Research, working on visual foundational models for autonomous driving
(2023-2024).
Previously, I was a research intern at WeRide,
Honda Research Institute, and Qualcomm AI
Research.
I sincerely appreciate the fellowship support from WeRide
(thank you! Dr. Yan Li, Dr. Hua Zhong, and Dr. Tony Han)
for funding my research in my first two years of study!
I am actively looking for an industrial RS or a postdoctoral
position! Please reach out to me if you think I might be a good fit!
google
scholar   |  
twitter   
|
|
News
-
[Sep 2024] Three papers accepted by CoRL! One on
using VLM to handle long-tail events in autonomous driving,
one on system-level failure detection for motion prediction model refinement,
and one on MoE policies for robot manipulation.
-
[Jun 2024] I was named a Rising Star and won the
Yunfan Award at the World Artificial
Intelligence Conference this year (top 15 early career Chinese AI researchers)!
-
[May 2024] I was named a Robotics: Science and Systems
Pioneer this year (top 30 early career researchers around the world in the robotics field)!
-
[Apr 2024] I won the
2024 Qualcomm Innovation Fellowship (32 winners in North America)!
-
[Mar 2024] I won the finalist award in the Baidu AI Fellowship this year (top 20 Phd students in
the AI field around the world)!
-
[Mar 2024] Our work RT-X won the 2024 Finalists for ICRA Best
Paper, Best Student Paper, and Best Manipulation Paper Awards
-
[Mar 2024] Check out our new preprint on a general calibrated regret metric for
detecting and mitigating system-level human-robot interaction failures and how it can be
used to identify informative deployment data for efficiently improving behavior prediction models.
-
[Jan 2024] Our work on visual representation
alignment for robot learning was accepted to ICLR 2024! We propose a tractable video-only method
for solving the visual representation alignment problem and learning visual robot rewards. Check
out the new results in the arXiv version.
|
Publications
For the most up-to-date list of publications, please see google scholar.
* indicates equal contribution and co-authorship.
|
|
Tokenize the World into Object-level Knowledge to Address Long-tail Events in Autonomous
Driving
Ran Tian, Boyi Li, Xinshuo Weng, Yuxiao Chen, Edward Schmerling, Yue Wang, Boris Ivanovic, Marco
Pavone
CoRL, 2024
paper
 
website
|
|
MEReQ: Max-Ent Residual-Q Inverse RL for Sample-Efficient Alignment from Intervention
Yuxin Chen, Chen Tang, Chenran Li, Ran Tian, Peter Stone, Masayoshi Tomizuka, Wei Zhan
preprint, 2024
paper
 
|
|
Not All Errors Are Made Equal: A Regret Metric for Detecting System-level Trajectory Prediction
Failures
Kensuke Nakamura, Ran Tian, Andrea Bajcsy
CoRL, 2024
paper
 
website
|
|
What Matters to You? Towards Visual Representation Alignment for Robot Learning
Ran Tian, Chenfeng Xu, Masayoshi Tomizuka, Jitendra Malik, Andrea Bajcsy
International Conference on Learning Representations, 2024
paper
 
|
|
Open X-Embodiment: Robotic Learning Datasets and RT-X Models
Google, Ran Tian, et al.
International Conference on Robotics and Automation (ICRA), 2024,
best paper award.
paper
 
website
|
|
Human-oriented Representation Learning for Robotic Manipulation
Mingxiao Huo, Mingyu Ding, Chenfeng Xu, Ran Tian, Xinghao Zhu, Yao Mu Lingfeng Sun, Masayoshi Tomizuka,
Wei Zhan
Robotics: Science and Systems, 2024
paper
 
website
|
|
Towards Modeling and Influencing the Dynamics of Human Learning
Ran Tian, Masayoshi Tomizuka, Anca Dragan, Andrea Bajcsy
International Conference on Human-Robot Interaction (HRI), 2023
paper
 
talk
|
|
Safety Assurances for Human-Robot Interaction via Confidence-aware Game-theoretic Human Models
Ran Tian, Liting Sun, Andrea Bajcsy, Masayoshi Tomizuka, Anca Dragan
International Conference on Robotics and Automation (ICRA), 2022
paper
 
talk
|
website adapted from here
|
|