We are steadily moving towards a future where humans work with robotic assistants, teammates, and even trainers. Towards realizing this future and mitigating its adverse side effects, I will share two techniques that together enable humans and robots to predict the other’s task-oriented behavior. First, I will present AI Teacher: an interactive machine teaching framework to assist humans in acquiring mental models of robots. By building upon human’s natural ability to model others (Theory of Mind), the AI Teacher framework reduces the number of interactions it takes for humans to arrive at predictive models of robot behavior. Second, I will discuss BTIL, an imitation learning approach for enabling robots to arrive at predictive models of humans’ collaborative behavior. Discussions of both these techniques will highlight the need as well as solutions for sample-efficient learning in settings of human-robot collaboration.
Vaibhav Unhelkar is an Assistant Professor of Computer Science at Rice University, where he leads a research group in the emerging area of Human-Centered AI. Ongoing research in his group includes development of algorithms and systems that enable “robots to work with humans” and “humans to become informed users of robots.” Unhelkar received his doctorate in Autonomous Systems at MIT (2020) and completed his undergraduate education at IIT Bombay (2012). He serves as an Associate Editor for IEEE Robotics and Automation Letters and is the recipient of JPMC AI Early Career Researcher Award and AAMAS 2022 Best PC Member Award. Before joining Rice, Unhelkar worked as a robotics researcher at X, the Moonshot Factory (formerly, Google X).