This week, Professor Lerrel Pinto spoke at the Edward J. Bloustein School of Planning and Public Policy, addressing the challenges faced in robotic learning. The talk, titled “Robot Data is Not Enough Data,” examined current limitations in how robots are trained to perform physical tasks.
Pinto noted that progress in robot learning has depended on large amounts of data gathered from humans operating robots. He argued that this approach has reached its limits because collecting robot data is expensive, slow, and limited in scope. He also pointed out that researchers do not yet know which demonstrations or labels are most important for developing intelligent machines.
“Robot data alone will never deliver the leap we need. We must demand more. Robots should learn directly from humans. They should feel the world through touch, rather than staring at pixels alone. And they must go beyond purely reactive modes and instead reason, plan, and act with foresight. If we are serious about building intelligent machines, we must move beyond the fixation on ‘just more data’ and instead embrace the hard, messy, human-centered problems that will define the next era of robotics,” Pinto said during his presentation.
The Bloustein School hosted Pinto as part of its ongoing commitment to advancing research in fields such as community development, transportation, health policy, workforce development, and energy policy through its various research centers (https://bloustein.rutgers.edu/). The school operates within Rutgers University and aims to foster inclusive and sustainable communities locally and globally (https://bloustein.rutgers.edu/).
Since 2013, the Bloustein School has recognized outstanding alumni through a Hall of Fame program as well as annual achievement awards dating back to 1994 (https://bloustein.rutgers.edu/). The institution’s graduate urban planning program ranks third nationally while its undergraduate public health program holds fourth place (https://bloustein.rutgers.edu/). Stuart Shapiro became dean of the school in 2023 (https://bloustein.rutgers.edu/).
For those interested in seeing examples of robots learning from humans or reading Dr. Pinto’s research papers, more information can be found on his website at https://www.lerrelpinto.com/.