Event
Microsoft Future Leaders in Robotics and AI Seminar Series: Suresh Kumaar Jayaraman
Friday, April 12, 2024
2:00 p.m.
Online Seminar
Enhancing Transparency in Human-Robot Teams: A Machine Teaching Approach to Communicate Robot Decision-Making to Diverse Human Teammates
Suresh Kumaar Jayaraman
Postdoctoral Fellow
Carnegie Mellon University
Absract
For transparent and effective collaboration between an agent and a human group, the challenge arises in teaching a diverse group of individuals about the agent’s decision-making process in a potentially time-sensitive and resource-limited environment. To address this challenge, we employ machine teaching through demonstrations for teaching groups of human learners modeling them as inverse reinforcement learners and using counterfactual reasoning to generate personalized informative demonstrations. Differing individual knowledge introduces challenges in personalization which we address by using aggregated team knowledge representations and developing models of team beliefs using particle filters. We present several group teaching strategies based on individual and aggregated team knowledge and conducted a simulation study comparing these different group teaching strategies with a baseline method of teaching each group member individually. We ran this study for various groups with varying combinations of learning abilities of its members. We discuss the results on multiple metrics such as learning resource utilized, learning rate, and final knowledge gained and discuss the implications of this work for human-agent teaming.
Bio
Dr. Suresh Kumaar Jayaraman is currently a postdoctoral researcher at the Robotics Institute at Carnegie Mellon University. He attained his M.S. and Ph.D. degrees from the University of Michigan in 2018 and 2021, respectively, following his B.E. from Anna University, India, in 2013. Before joining CMU, he held a postdoctoral position at Cornell University. Dr. Jayaraman's research interests revolve around the development of trustworthy robots for seamless human-robot collaboration and teaming. His current focus lies in group human-robot interaction, with an emphasis on explainability, implicit communication, and fostering trust in robots within groups. His research methodology employs a multifaceted approach. Firstly, he delves into understanding how robot communication influences explainability and trust, particularly exploring the nuances of implicit communication conveyed through robot movements and demonstrations. Secondly, he is involved in developing computational models that depict human learning and behavioral responses when interacting with robots. Lastly, he is dedicated to crafting adaptive robot algorithms that enhance explainability and sustain trust in robots over time.
