Social Influence Games
Modeling opinion dynamics under competing external influence operations.
Bio
Nandan Tumu is a PhD Candidate at the University of Pennsylvania. His research focuses on developing trustworthy learning methods for robotics. His work sits at the intersection of uncertainty quantification, physics-informed machine learning, and multi-agent decision-making. He aims to improve Trustworthy AI by integrating inductive biases informed by dynamics or physics, using finite-sample methods of uncertainty quantification, and exploiting the structure of inter-agent interaction. These methods improve the reliability of learning-enabled robotic systems operating in novel and adversarial environments. His research is supported by the NSF Graduate Research Fellowship.