Events
Tuesday, October 7th, 2025 - Texas Robotics Talk
Speaker: Yang Gao, Assistant Professor, Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University, and Co-founder of Spirit.AI
Talk title: Manipulation Data Pyramid: From Human Video Pretraining to Physical RL
Abstract: Scaling laws are increasingly viewed as a cornerstone on the path toward general intelligence. In robotics, however, progress is often hindered by a lack of abundant, high-quality data. In this talk, Dr. Yang Gao introduces a data pyramid strategy designed to overcome this challenge by maximizing the potential of diverse data sources.
The approach combines internet-scale datasets, human teleoperation data, and robot-collected experiences—each reinforcing and filling the gaps of the others.
Dr. Gao will illustrate this strategy through three examples:
A system that learns directly from human VR recordings, enabling zero-shot motion generalization.
A study demonstrating how data scaling laws emerge in imitation learning with manually collected demonstrations.
A new reinforcement learning framework that leverages foundation models to accelerate learning in the physical world.
The central message: data is the wellspring of knowledge. By shifting focus from purely algorithmic design to strategically building and utilizing data, we can begin addressing some of the biggest challenges on the road to general robotic intelligence.
About the speaker: Yang Gao is an Assistant Professor at the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University, and co-founder of Spirit.AI, a company focused on building general-purpose robots. He received his Ph.D. from UC Berkeley under the supervision of Prof. Trevor Darrell, followed by postdoctoral research with Prof. Darrell and Prof. Pieter Abbeel.
His research centers on robotic learning. Before Berkeley, he earned his bachelor’s degree in computer science from Tsinghua University, where he worked with Prof. Jun Zhu on Bayesian inference. He also explored autonomous driving during a research internship at Intel Labs in 2018 with Dr. Vladlen Koltun.
Date: Tuesday, October 7, 2025
Time: 11:00am - 12:00pm
Location: AHG 1.112 | Zoom
Minor in Robotics Information Session
Info Session #1:
Date: Tuesday, October 7, 2025
Time: 5:00pm - 6:00pm
Location: AHG 1.112
Info Session #2:
Date: Wednesday, October 8, 2025
Time: 12:00pm - 1:00pm
Location: AHG 1.112
Info Session #3:
Date: Spring 2026
Time: TBD
Location: TBD
Info Session #4:
Date: Spring 2026
Time: TBD
Location: TBD
Thursday, October 9th, 2025 - National Fellowship Q&A + Mentoring
Join Texas Robotics faculty and graduate student winners of national fellowships for a special Q&A session on competitive graduate research fellowships. Breakfast will be provided.
After the session: sign up for 1-on-1 peer mentoring on your proposal in the days leading up to the deadlines.
Date: Thursday, October 9, 2025
Time: 9:00am - 10:00am
Location: AHG 1.112
Every Other Tuesday (Starting Sept. 2) - Texas Robotics Portfolio Seminar Series
The Portfolio Seminar Series is an event for Texas Robotics Graduate Students at the University of Texas at Austin. It is held online and in person at 12:00pm every other Tuesday.
Texas Robotics Graduate Students Meetings
TRGS meetings are held every Monday at 1pm in the conference room of AHG (2.124). Attendance and participation in the organization is encouraged for all graduate students affiliated with Texas Robotics.
By entering an event or program of the Texas Robotics, you are entering an area where photography, audio and video recording may occur.
Your entry and presence on the event premises constitutes your consent to be photographed, filmed, and/or otherwise recorded and to the release, publication, exhibition, or reproduction of any and all recorded media of your appearance, voice, and name for any purpose whatsoever in perpetuity in connection with the Texas Robotics and its initiatives, including, by way of example only, use on websites, in social media, news and advertising.
By entering the event premises, you waive and release any claims you may have related to the use of recorded media of you at the event, including, without limitation, any right to inspect or approve the photo, video or audio recording of you, any claims for invasion of privacy, violation of the right of publicity, defamation, and copyright infringement or for any fees for use of such record media.
You understand that all photography, filming and/or recording will be done in reliance on this consent. If you do not agree to the foregoing, please contact robotics@utexas.edu.
