Lehigh to host NSF-supported workshop in October to help researchers help robots learn

Robots are everywhere. In fact, every time you brake your car, a robot is interpreting that pressure signal from your foot while simultaneously assessing the conditions of the road. Based on those two inputs, the robot decides if the wheels will stop turning immediately, or if the antilock system needs to kick on so you don’t skid out on a slippery street.

Programming robots to perform tasks is difficult. Motivated in part by this challenge, researchers are devising algorithms that allow robots to learn and adapt quickly to their environments, and they are currently investigating two sides of the machine learning problem—one in which the robots have too many examples to learn from, and one in which they have too few. 

To help solve these problems and foster interdisciplinary communication and collaboration around this emerging field, Lehigh University will host a two-day Robot Learning Workshop October 14-15.

The event is the second in a series of conferences and lectures funded by a National Science Foundation TRIPODS-X grant awarded to Lehigh’s Institute for Data, Intelligent Systems, and Computation (I-DISC). The series addresses topics including applications of machine learning and big-data science in chemical processes, autonomous robotics, supply chain optimization, and cognitive neuroscience. 

The October workshop will bring together leading researchers in robotics and offer numerous presentations on emerging directions at the intersection of robotics, deep and reinforcement learning, control systems, and operational research. By facilitating interactions between researchers across disciplines who are interested in designing and implementing autonomous robots that can learn to perform their tasks, organizers hope the event will jump-start work in new interdisciplinary directions. 

“There is a clear need for developing algorithms that are able to learn automatically and have the robot adapt to the environment,” says I-DISC co-director Héctor Muñoz-Avila, a professor of Computer Science and Engineering at the P.C. Rossin College of Engineering and Applied Science and one of the workshop organizers. “There is consensus in the research community about that—one size does not fit all. So anyone who is interested in robotics will get something out of this. I’m very excited about it, as we have an outstanding group of 16 speakers.”

The event will take place in Iacocca Hall on Lehigh’s Mountaintop Campus in Bethlehem, PA, and will include a poster session and reception. The cost to register is $50 (waived for Lehigh faculty and students), and a limited number of travel grants are available for students interested in attending the workshop and presenting a poster. The deadline to register is October 1.

The third workshop in the series, scheduled for spring 2020, will focus on cognitive neuroscience.

Robot learning

“Robot Learning Workshop” will feature presentations on emerging directions at the intersection of robotics, deep and reinforcement learning, control systems, and operational research.

Iacocca Hall

The October workshop will be held in Iacocca Hall on Lehigh University's Mountaintop Campus.

Hector Muñoz-Avila

Hector Muñoz-Avila, I-DISC co-director and a professor of computer science and engineering