Cooperative Multi-Agent Plan Recognition: A Review of the State of the Art by Loren M. Champlin
Bio: Loren M. Champlin is a Ph.D. student at the School of Information minoring in Cognitive Science while also completing a graduate certificate in Computational Social Sciences at the University of Arizona. He has also completed a B.S. in Applied Mathematics with a minor in Computer Science, a M.Ed. in Teaching and Teacher Education, and a M.S. in Statistics all from the University of Arizona. Currently, his main research focus is Symbolic Artificial Intelligence, specifically the topics of automated planning, plan recognition, and knowledge engineering as they relate to Cognitive Science and Theory of Mind. His current work focuses on developing an AI system that is capable of understanding and assisting teams of humans cooperating on a specific goal or task.
Abstract: There is an increasing need for technologies that can assist groups of human users cooperating on a coordinated task. An essential component to providing this need is an artificial intelligence (AI) system that can recognize and understand the behaviors and plans of a cooperating group of human users. This capability is known as Plan Recognition or more specifically Cooperative Multi-Agent Plan Recognition (MAPR) when dealing with multiple cooperating agents. My talk introduces Plan Recognition and Cooperative MAPR, describes the taxonomy of approaches for implementing Cooperative MAPR, and discusses some conclusions I have made about the state of the field.