Robert is a first year PhD fellow at the Einstein Center for Neurosciences Berlin. Under the supervision of Henning Sprekeler, he investigates the learning mechanisms underlying intelligent swarm behavior. By combining Deep Multi-Agent Reinforcement Learning as well as evidence from animal psychology, he aspires to solve the inter-temporal as well as inter-agent credit assignment problem. Before joining the lab in May 2019, Robert completed a MSc in Computing (Machine Learning) at Imperial College London, a Data Science MSc at Universitat Pompeu Fabra and an Economics undergrad at the University of Cologne. He enjoys long walks, ice cream, open source projects as well computational arts. And from time to time he writes down his thoughts.
Learning of Intelligent Swarm Behavior
Collective behavior is puzzling and fascinating at the same time. Until now the emergence of such large-scale dynamics has mainly been studied from an evolutionary perspective. In this project, on the other hand, we investigate how normative rewards and top-down learning give rise to intelligent social behavior. More specifically, the goal of the project is to scale Deep Multi-Agent Reinforcement Learning to swarm systems and to provide a set of solutions to the highly non-stationary learning process. By understanding the fundamental mechanisms underlying collective adaptation, we envision the design of artificial shepherds that are able to guide the collective’s inter-temporal decision making process. Thereby, we want to escape sub-optimal equilibria and make the world a better place.